Telecommunications and Networks

By the same Author: Network Analysis, Prentice Hall, 1969 Development of Information Systems for Education, Prentice-Hall, 1973 Information Processing Systems for Management, Richard Irwin, 1981 and 1985 Information Resource Management, Richard Irwin, 1984 and 1988 The Computer Challenge: Technology, Applications and Social Implications, Macmillan, 1986 Information Systems for Business, Prentice Hall, 1991 and 1995 Management of Information, Prentice Hall, 1992 Artificial Intelligence and Business Mangement, Ablex, 1992 Knowledge-Based Information Systems, McGraw-Hill, 1995 Managing Information Technology, Butterworth-Heinemann, 1997

Telecommunications and Networks

K.M. Hussain D.S. Hussain

Butterworth-Heinemann Linacre House, Jordan Hill, Oxford OX2 8DP A division of Reed Educational and Professional Publishing Ltd A member of the Reed Elsevier plc group

First published 1997 © K. M. and Donna S. Hussain 1997
All rights reserved. No part of this publication may be reproduced in any material form (including photocopying or storing in any medium by electronic means and whether or not transiently or incidentally to some other use of this publication) without the written permission of the copyright holders except in accordance with the provisions of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London, England W1P 9HE. Applications for the copyright holder’s written permission to reproduce any part of this publication should be addressed to the publishers

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. ISBN 0 7506 2339X Library of Congress Cataloguing in Publication Data A catalogue record for this book is available from the Library of Congress.

Typeset by Laser Words, Madras, India Printed in Great Britain by Martius of Berwick

bridge. Japan Case 1.1: Network disaster at Kobe. repeater and gateway Compression Addressing Modems Smart and intelligent xii 1 1 3 6 10 10 10 11 13 15 15 17 20 21 22 23 24 24 25 26 26 26 28 28 29 32 33 35 35 36 36 38 38 39 40 41 42 43 v 3 4 .1: The Iridium project Case 3.3: Transmission at a trillion bits per second Bibliography Switching and related technologies Introduction Router.1: Milestones for network development Bibliography PART 1: TECHNOLOGY 2 Teleprocessing and networks The rise of distributed data processing Transmission channels Interconnectivity Networks in the 1990s Issues facing corporate management Summary and conclusions Case 2.1: Delays at the Denver Airport Supplement 2.2: CT-2 and PCs in the UK Case 3.CONTENTS Acknowledgements 1 Introduction Changes in technology Management of telecommunications Applications of telecommunications Case 1.1: Top telecommunications companies in 1994 Bibliography Transmission technologies Introduction Wiring Microwave Satellite Wire-less/cordless systems Comparison of transmission systems Summary and conclusions Case 3.2: Networking at the space centre Supplement 1.

1: ATM at sandia Case 6.2: Survey on WANS Supplement 6.1: Networking at the space centre Bibliography 5 LANs: local area networks Introduction Interconnectivity Characteristics of networks Networking as a computing paradigm Topologies and switches Access methods Ethernet The token ring Circuit switching FDDI Frame relay SONET Wire-less networks Summary and conclusions Case 5.2: ISDN in France Case 7.Contents Protocols Hubs Summary and conclusions Case 4.2: The Stentor network in Canada Case 5.1: A network in the UK Case 5.3: Projected pricing of ATM Bibliography ISDN Introduction The computing environment The resource environment What is ISDN? Implementation of ISDN Summary and conclusions Case 7.1: ISDN at West Virginia University (WVU) Case 7.2: Navigating LANs/WANs in the UK Supplement 6.3: ISDN for competitive bridge across the Atlantic Bibliography 45 45 46 47 47 48 48 48 50 50 52 55 55 55 55 56 56 56 56 57 57 57 58 58 59 59 59 60 61 61 62 62 65 66 67 67 67 68 68 69 69 69 72 72 73 75 76 76 77 77 6 7 vi .1: Wan technologies Supplement 6.3: A network planned for China Bibliography MAN/WAN Introduction MAN and WAN Planning for a WAN Performance of a MAN/WAN Bandwidth management Switching management The ATM Summary and conclusions Case 6.

2: Networking standards in Europe Bibliography Security for telecommunication Introduction 78 78 78 80 82 83 84 85 87 87 87 88 89 91 91 91 93 95 95 96 97 98 99 99 99 100 100 100 106 108 108 109 110 110 111 111 112 112 112 116 117 118 120 120 121 122 122 123 123 vii 10 11 12 .3: Networking at SKF. Sweden Bibliography PART 2: ORGANIZATION FOR TELECOMMUNICATIONS AND NETWORKS 9 Organization for networking Introduction Location and organization of network management Structure of network administration Planning for networking Planning variables Dynamics of network planning Planning process Implementing a plan Summary and conclusions Case 9.3: Applications of client server systems Bibliography Standards Introduction What are standards? The development of OSI The ISO European standards organizations TTC in JAPAN Summary and conclusions Case 11.1: AAPN in HFC Bank Case 8.2: Hidden costs of APPN Case 8.Contents 8 Network systems architecture Introduction Systems network architecture (SNA) The OSI model The APPN TCP/IP Multiple protocols Summary and conclusions Case 8.1: Development of international standards for the B-ISDN in the US Case 11.1: Headaches for network management Bibliography The client server paradigm Introduction Components and functions of a client server system Organizational impact Advantages of the client server system Obstacles for a client server system Summary and conclusions Case 10.2: Citibank’s overseas operations in Europe Case 10.1: Client server at the 1994 Winter Olympics Case 10.

2: Examples of malicious damage Case 12.1: Popular viruses Bibliography 13 Network management Introduction Management of networks Software for network management User management Development of networks Resources for network management Summary and conclusions Case 13.1: Milestones towards the development of an NII Bibliography Global networks Introduction 124 124 126 128 129 130 132 133 134 135 137 137 137 138 138 138 139 139 140 140 141 144 144 145 147 151 151 151 152 152 153 153 153 155 158 159 160 160 161 161 162 163 167 167 168 168 168 170 170 171 171 14 15 16 viii .3: German hacker invades US defence files Case 12.1: Prices of LAN management software in 1995 Bibliography Resources for teleprocessing Introduction Parallel processing Software for telecommunications Summary and conclusions Case 14.6: Miscellaneous cases using telecommunications Supplement 12.5: The computer ‘bad boy’ nabbed by the FBI Case 12.Contents Security Terminal use controls Authorization controls Communications security Security for advanced technology Computer viruses Policies for security Administration of authorization How much security? Summary and conclusions Case 12.4: Buying the silence of computer criminals Case 12.1: Alliances and mergers between carriers Case 15.1: Networking in the British parliament Case 13.3: Telecommunications law in the US Supplement 15.1: Examples of hacking Case 12.2: Share of the European VAN market Case 15.2: Analyser at Honda auto plant Supplement 13.1: Replacement of mainframes at EEI Supplement 14.1: Top world telecommunications equipment manufacturers in 1994 Bibliography National information infrastructure Introduction NIIs around the world NII in the US Issues for NII Summary and conclusions Case 15.

3: Telecom leap-frogging in developing countries Case 16.1: GE bases global network on teleconferencing Case 17.2: Index of global competitiveness Supplement 16.2: Electronic data exchange (EDI) in the UK Supplement 17. German and US companies Supplement 16.3: Telecommunications media for selected countries in 1994 Supplement 16.4: Telecommunications end-user service available in regions of the world Bibliography PART 3: IMPLICATIONS OF NETWORKS 17 Messaging and related applications Introduction Teleconferencing Electronic data interchange (EDI) Standardization Electronic transfer of funds EFT spin-offs Cooperative processing Message handling systems (MHS) Summary and conclusions Case 17.1: Global outsourcing at Amadeus Case 16.2: Telstra in Australia Case 16.4: Slouching towards a global network Case 16.5: Alliance between French.1: World-wide software piracy in 1994 Supplement 16.1: Costs of message handling and related processing Bibliography Multimedia with telecommunications Introduction Multimedia and distributed multimedia Requirements of multimedia Resources needed for multimedia processing Servers Networks Clients Standards Applications of distributed multimedia Video-conferencing Video/film-on-demand Telemedicine Digital library Distance learning Multimedia electronic publishing Organizational implications of distributed multimedia 172 173 174 175 176 178 179 180 181 181 182 182 183 183 183 183 184 184 185 187 187 188 188 189 190 192 194 195 196 197 198 198 198 199 199 199 200 201 202 203 203 204 204 204 206 206 207 208 209 210 ix 18 .Contents Global networks Consequences of global networks Telecommunications and developing countries Global outsourcing Transborder flow Protection of intellectual property Global network and business Summary and conclusions Case 16.

3: Business on the Internet Case 20. e-mail and information services Introduction Telecommuting Implementation of teleworking The benefits and limitations of telecommuting When telecommuting? Future of teleworkers E-mail Resources for e-mail Information service providers Summary and conclusions Case 19.3: The access projects at the British Library Case 18.6: Users of the Web for business 211 213 213 214 214 214 215 216 216 216 217 219 220 220 221 223 225 225 227 227 228 228 228 228 229 229 229 231 231 232 232 232 234 236 236 240 241 241 242 243 244 245 245 245 246 246 246 246 246 247 247 20 x .4: Advice from teleworkers Case 19.6: Internet in Singapore Case 20.5: Telemedicine in Kansas Bibliography 19 Telecommuters.4: Video-conferencing in telemedicine at Berlin Case 18.5: Court in France defied in cyberspace Case 20.1: Growth of the Internet Supplement 20.1: Intrusions into Internet Case 20.5: Teleworking at AT&T Case 19.2: Growth in Internet hosts around the world Supplement 20.2: Comments from teleworkers Case 19.3: Telecommuting at American Express Case 19.4: Users of the Internet in 1994 Supplement 20.2: Bits and bytes from cyberspace Case 20.Contents Summary and conclusions Case 18.6: Teleworking in Europe Case 19.5: Build or rent a Web site? Supplement 20.3: Computers connected to the Internet in 1994 Supplement 20.1: Examples of teleworkers Case 19.2: Electronic publishing at Britannica Case 18.7: Telecommuting in the US (in 1994) Case 19.8: Holiday cheer by electronic mail Bibliography Internet and cyberspace Introduction Cyberspace Internet Connecting to the Internet Surfing on the Internet Internet and businesses Security on the Internet Organization of the Internet The Internet and information services Summary and conclusions Case 20.7: English as a lingua franca for computing? Supplement 20.1: MedNet Case 18.4: Home-page for Hersheys Case 20.

1: Percentage growth of phone lines connected to digital exchanges in the 1990s in Europe Supplement 21.7: Milestones in the life of Internet Bibliography 21 What lies ahead? The future in the context of the past Trends in telecommunications technology Network management and the future Management of the Internet Standards A t´l´matique society ee Advanced applications Predictions often go wrong! Summary and conclusions Case 21.3: Minitel its past and its future Supplement 21.2: Spamming on the Internet Case 21.Contents Supplement 20.1: Amsterdam’s digital city Case 21.2: World-wide predications for 2010 compared to 1994 Bibliography Glossary of acronyms and terms in telecommunications and networking Acronyms in telecommunications and networking Glossary Index 247 247 249 250 251 255 255 257 257 259 261 263 265 265 266 267 267 267 269 270 272 277 xi .

ACKNOWLEDGEMENTS The authors wish to thank colleagues for their helpful comments and corrections to the manuscript. Thanks to Tahira Hussain for her help. Frank Leonard. Chetan Shankar and Derek Partridge. especially in the preparation of the diagrams using the PowerPoint program. . Any errors that still remain are the responsibility of the authors. These include Linda Johnson.

For long distances. Dertouzos. Early transmission was by wire. we needed data communications to transmit data to remote points. This extended use of telecommunications is the subject of this book. Telephones and telegraph were complemented by post and organized as a utility better known as the PT&T (Post Telephone and Telegraph). radio broadcasts and satellite are superior. with Germany and Japan in the lead and the UK and USA not far behind. Its advantages are.75 million pages of double spaced text. Instead. 1 . This will provide us with a framework in which we can then place the many components of the technology of telecommunications and networks. the equivalent of 18. a ‘soft’ subject that we chose to avoid here. Later. With the increased volume and complexity of messages now being sent. and even images and video. on engines and fuel that fed them. A typical fibre optic cable can carry up to 32 000 long distance telephone calls at once. which uses thin glass fibres that are both cheaper and less scarce than copper. Bell Labs developed the rainbow technology that sent three billion bits of information down one fibre optic thread in one second. the industrial age. In the USA. Soon after. But from the broadcasting and satellite station. restricted to the distance transmitted. Fibre is also less bulky than copper and much lighter.1 INTRODUCTION The agricultural age was based on ploughs and the animals that pulled them. the equivalent of 2. these services have been privatized and other countries may follow the path away from monopoly towards privatization and free competition. these points of communication increased in number. One strand of fibre thinner than a human hair can carry more messages than a thick copper cable. the need for fibre is great and no longer in dispute. the connection of remote points by telecommunication is referred to as a network. In the USA and UK. But this is a controversial question of politics and government policy-making. The earliest transmissions were by telephone for voice and telegraph for the written word. voice. Michael L. 1991 Telecommunications is an old and stable technology if you think only of telephones and telegraph. Many countries are turning to fibre. Changes in technology We start with an overview of the technology in Chapter 2. We shall examine the technology in the first part of the book. however. copper wires to be more precise.5 billion bits of data per second. we will confine ourselves to the more ‘hard’ and stable topics of technology: the management and applications of the technology. It has been replaced by fibre optics. But copper is both expensive (and sometimes scarce) and bulky. the use of fibre for data communications has risen 500% during the period 1985 90. Fibre optics will be used for short distance transmission and will complement radio broadcasting and satellite. The information age we are now creating will be based on computers and networks that interconnect them. Fibre optics is less expensive than stringing wire across telephone poles and even less expensive in capital cost than cable. The first of these technologies to be examined is transmission. with the transmission no longer being limited to data but included text. the connection to the home or the office must still be made by wire or by fibre. But then in the 1960s came computers and the processing of data. Recently. the management of the telecommunications in the second part. and in the third and final part the many applications that are now possible because of telecommunications. Back to transmission.

a device that translates from an analogue signal of. The technologies described above are fairly stable and have well established standards that 2 . but transmission needs a modem. Eliminating these redundancies and compressing the message into a smaller sized message without losing any content is called compression. One device that determines the route (path) that a message takes across switches. Many messages tend to have redundancies and even blanks (as in the sentences of this book). Both ISDN and B-ISDN are the subject of Chapter 8. The earliest network was implemented by the US Department of Defense to facilitate the communication between their researchers and academics working on defence projects. Transmission is just one of the technologies enabling telecommunications. say. bridges (and/or) gateways is the router. waiting at the airport. An international standard and one that is accepted globally is the ISDN (Integrated Systems Digital Network). and is easier and cheaper to maintain and operate. as reflected in the expenditures on ISDN which have doubled in the last three years since 1994. Conversion to ISDN is expensive and slow but steady in the US.1. These researchers (and later others in private Wireless Fibre Satellite Radio Copper Wire / Cable Figure 1. We have mentioned networks as being interconnected points of communication. a telephone to a digital signal of a computer and vice versa. Other technologies include the many devices that make telecommunications possible by contributing to the transport of messages over networks. instead of the two signals (analogue and digital) that we now carry. A description of the OSI and its competitors in the US (the SNA and the TCP/IP) are examined in Chapter 7. This evolution of transmission is the subject of Chapter 3 and is summarized in the spiral of change shown in Figure 1. which is now in the stages of getting international standards. They unknowingly sowed the seeds of e-mail (electronic mail) and many other applications of telecommunications. The evolution of these enabling technologies is shown in Figure 1. The demand for services to be transmitted has resulted in plans to extend ISDN to B-ISDN (Broad-band ISDN). Integrated digital transmission is faster. These individuals were technical and inquisitive and became interested in developing a more reliable and efficient way of communicating not just their research projects but everything else including their daily mail. which are now carried on telephone lines. This makes the transmission efficient by taking less space (and time) to transmit the message. or even while walking the dog. They make transmission so much more portable that one can now transmit while driving a car. but the efficiency of the transmission depends largely on the message being transported. ISDN will enable the transmission of analogue signals. This is the OSI (Open Systems Interconnection) model that was designed as a framework for the structure of telecommunications and networking. digital. emerging strongly is the demand for wireless or cellular phones. These devices are the subject of Chapter 4. as digital signals. This device contributes to the effectiveness of the transportation of the message.Telecommunications and networks are universally accepted.1 Evolution of transmission media However. One technology that does not have universal acceptance and is very controversial is the international standard for an architecture and protocol for networks that is open to varying designs of hardware and software. One set of such devices include the bridge that connects homogeneous (similar) networks and the gateway that connects non-homogeneous (dissimilar) networks. This conversion of the analogue world to the digital world has already resulted in the infrastructure becoming overloaded and overwhelmed. This enables us to have just one signal. requiring equipment for interfacing and resulting in both inefficiencies and high cost.2. like those used by the common desktop computer.

B-ISDN OSI / SNA / TCP/IP Smart devices Bridges / Gateways Compression Modems Figure 1. In ten years. The MAN and the WAN are the subject of Chapter 6. 5) Figure 1.Introduction ISDN.2 Spiral of enabling technologies 1074 interconnected computers. With large volumes of data. it has become a very effective and popular means of communication. 6) LAN (Local Area Network) (Ch. Despite no formal initiation or structure. industry) were also interested in developing a worldwide network of communications. In 1984. and in a yet wider network the WAN (Wide Area Network). The Internet is now being used not just by researchers but by individuals and also by businesses. Along the way to the Internet. The Internet is a network of other networks. The Internet is discussed as an application of telecommunications in Part 3. the number grew to 3. These networks use architectures and protocols discussed in Chapter 8 and may or may not use the ISDN examined in Chapter 7. the Internet had Management of telecommunications The technologies mentioned above (and defined operationally) are all examined in detail in S M D S (Switched Multimegabit Data Service) G A N (Global Area Network) (Ch. 6) Internet (Ch. more specifically in Chapter 20.3 Spiral of networks 3 . Their evolution is shown in Figure 1. Networking in a broader geographic area is referred to as MAN (Metropolitan Area Network). 6) MAN (Metropolitian Area Network) (Ch.8 million and is still growing.3. and so ARPANET and the many technologies that it developed eventually led to the Internet. the ARPANET contributed to the evolution of formal networking first in small local areas better known as the LAN (Local Area Network). 5) ARPANET (Ch. 20) WAN (Wide Area Network) (Ch. systems like the SMDS (Switched Multimegabit Data Service) will become more common in the future. This is the subject of Chapter 5.

9) Decentralized (Ch. Distributed Data Processing. 10) Distributed (Ch. 16) N I I (National Info. These organizational approaches are examined in Chapter 9. Infrastructure) (Ch. where the computer at a remote node is a client and the common computing resources (like data. 9) Centralized (Ch. the Personal Computer. The end-user (ultimate user of computer output) was becoming computer literate. 9) Figure 1. (At that time. Parallel to the growth of PCs was the dissatisfaction of the end-user of the centralized approach which was slow and unresponsive. The endusers now had the desire (and sometimes with a passion) for the control of local operations. They wanted the centre to do the planning of commonly needed resources (equipment. Thus evolved the client server system. This facilitated the economic use of the scarce resource of computer personnel as well as the expensive equipment.Telecommunications and networks Part 1 of this book. Infrastructre) (Ch. most of computer processing was for data and only later did it extend to text. voice. We start in Chapter 9 with the location and organization of telecommunications as part of IT (Information Technology) and as part of a corporate organization structure. The end-users were willing to accept many of the responsibilities of maintaining and even selecting resources and developing systems needed at the remote nodes. 15) Client−Server System (Ch.4 Spiral of organization for telecommunications 4 . and a parallel increase in the ability and desire for the centralized power to be decentralized to the remote nodes where the computing needs resided. video and images). and no longer cowed by the computer specialist at the centralized and remote location. PCs made distribution economically feasible and telecommunications which interconnected these nodes made it a feasible proposition. Part 2 is concerned with the management of these technologies. But at a national and G I I (Global Info. This led to DDP. These are identified and discussed in Chapter 10. Such a system requires solutions to special computing resources and the solution to many organizational and managerial issues. But then came the PC. databases and even technical human resources) and the development of mission critical applications whilst leaving the computing at the nodes to the end-user. The earliest organization structure was to centralize the large computer processors and mainframes that served all the local and remote users. The client server approach is appropriate for a corporation or institution. knowledge and application programs) reside on computers called servers.

but certainly not enough. This evolution in the organization of telecommunications is shown in Figure 1. There is more data (and computer programs) that can be accessed and there is also more money that can be transacted across the lines of telecommunication. If all of us were to pursue our own preferences in design and conventions for operations we would never be able to communicate with each other and there will be no compatibility and interoperability of devices and protocols (procedures). It came too late and had to face entrenched vested interests of manufacturers and suppliers of telecommunications equipment. The need for standards and the process of agreeing on standards by balancing the often conflicting interests is the subject of Chapter 11. The management of standards is the subject of Chapter 11.Introduction regional level an infrastructure for telecommunications is desirable that will not only meet the high demands of volume but the diverse demands of not just data but also voice and image. Standards are agreed upon conventions and rules of behaviour are part of our daily life and certainly not new to IT where we have standards for hardware and software and even standards for analysis and design. as with standards. The question for management (corporate and telecommunications) is not whether we need security but how much and where. What is new is the nature of resources that have to be acquired. But in telecommunications there are additional dimensions. Thus the task of telecommunications management is to correctly select the best technology and to assess the timing of adopting now and run the risk of being outdated or waiting and not benefit from existing advances in technology. For example. The international standards on network architecture mentioned earlier. In the case of global telecommunications this may be a continent away or across the oceans. Just as the infrastructure for road transportation changes from a city and local transportation to a motorway (freeway or autobahn) with all its interconnections. But the timing was wrong. But standards in high tech industries like telecommunications must not come too early before the technology is stabilized for that will ‘freeze’ and discourage newer approaches and innovations. Also. this is not new to IT management. In a client server environment. The carrying capacity has to increase. But there is the need to select the processors needed for accessing the network. It is important for telecommunications. one needs 64 000 bits per second capacity to transmit voice. Standards is one of the issues faced by management of telecommunications. the temptation is larger. We have all these types of standards for telecommunications plus a few more.2 million bits per second to transmit high fidelity music. The managerial issues of telecommunications is the subject of the following four chapters (Chapters 11 15). We do have agreement of many standards including international standards. Another concern of telecommunications management is security. 1. the OSI. and of security in Chapter 12. The process of acquisition is not new to either IT or to any corporation. Management must assess the cost of security and compare it with the risk of exposure. We do not have all the international standards we need. It is likely that your plug may not fit into the socket in the wall. Acquisition of telecommunications resources is the subject of Chapter 14. Chapter 13 is an overview of the management and administration of all of telecommunications and networking. encryption and other approaches are also needed to protect selected messages that are transmitted. The potential population of those who can penetrate the system is larger since there are now more people who have computers and know how to use them. Again. The acquisition and organization of telecommunication resources is covered in Chapter 14. One can experience that by going to another country and trying to plug in a computer. These subjects are discussed in Chapter 12. We thus need to control the access to networks by building fire-walls to protect our assets.4. the client may be 5 . if you consider the fact that telecommunications involves remotely located parties. They take a lot of effort and time. so also we need an entirely different set of transmission capacities and enabling technology for interconnectivity to connect and handle transmission. Such an infrastructure for national telecommunication (NII) is discussed in Chapter 15 and for a Global Information Infrastructure (GII) is discussed in Chapter 16. and 45 million bits per second to transmit video. At the corporate level the decision is that of selecting a LAN or MAN or WAN and not of selecting the devices for connectivity or the media of transmission which is part of the infrastructure. took ten years.

6. which is used extensively for transfer of documents and files by businesses. acquisition and maintenance of all the resources needed for telecommunications. school. A summary of these activities is shown in Figure 1. and a look into the future is Chapter 22. And so we need to consider not just file servers and application servers but also video servers. The server is a computer that could vary from a powerful PC to a mini or mainframe. integrated applications is Chapter 21. also in business but restricted to financial institutions . multimedia is Chapter 18. Other applications of message handling are not so conspicuous but just as important. medical facility or government agency. 13) Figure 1. the Internet in Chapter 20. We have evolved from the stand-alone computer system to a system with a variety of computers that serve as clients or servers or both and are interconnected by telecommunications. We need more 6 than standards for such integration and many of the issues that arise are not just technological but economic and political. office. 14) Planning OPERATIONS Development Administration and Management of Telecommunications (Ch.5 Management of telecommunications a PC or a workstation. 12) Resource Acquisition Mgmt. But the servers for tomorrow have to be capable of handling not just data but multimedia. A graphic summary of this flow of topics is shown in Figure 1. often called the information highway. This infrastructure. Electronic Data Interchange.Telecommunications and networks Standards Management (Ch. global outsourcing and the protection of intellectual property. These are examined in Chapter 15.5. These issues are examined in Chapter 16. Applications of telecommunications The next and final part of the book is concerned with applications of telecommunications and networks. It is concerned with the planning. An example of such late maturing applications is e-mail (electronic mail) that has suddenly ‘taken-off’ with high rates of growth and become a ‘killer’ application. The last two chapters on management of telecommunications (Chapters 15 and 16) go beyond the corporate level. This includes personnel resources discussed in Chapter 9 on the organization at the corporate level. business. One is EDI. Message handling is Chapter 17. 11) Security Management (Ch. Chapter 15 is concerned with integration at the national level by providing an infrastructure for telecommunications much like we have an infrastructure for communications by road or plane. provides the interconnections for exchange of information and enables the integration of all the sources and destinations of information whether this be the home. library. It is so important an application that it will be discussed at great length later in Chapter 19. Another application. (Ch. We compound all these problems when we consider global communications and have additional issues of transborder flow. Our first discussion of applications will be on message handling applications in Chapter 17. Some of these applications have been around for a long time. teleworking is Chapter 19. The chapter on management of telecommunications (Chapter 13) is more of a summary of all the related chapters. The importance of telecommunications management can be gauged by the statistic that corporate spending on telecommunications in the US has more than doubled in the three years since 1994.

news. In 1995. In 1995. In 1995 there were over eight million subscribers. Telemedicine could also be valuable as a first opinion for those who may be located remotely (permanently or temporarily as when travelling). and as such can reside in computer storage.. It started by renting computer time from an insurance company that had purchased a computer and had unexpected excess capacity. One service provider is CompuServe. It is much faster and more reliable than traditional mail. 7 . to be transferred to an expert anywhere in the world for a second opinion. are also becoming multimedia thereby greatly improving the quality of what is transmitted. as in many teleprocessing applications. Information services could be customized so that you select what you want from the diverse options and do not have to take what is edited and passed down as is the case with the 12 000 newspapers and magazines and the many TV stations. This may well affect our learning as well as our patterns of how we spend our leisure time. distance learning and electronic publishing. this may well evolve into videoconferencing. with payments consisting of data transfers between one machine and another. or browse through the contents of the Tate Gallery in London without having to physically visit the place. As James Martin once put it. One final application of multimedia to be discussed here briefly is the use of telecommunications in medicine. Electronic Funds Transfer. in just one country (US). delivered to the home at any time of the day or night. We take such electronic transactions for granted little realizing that if it were not for telecommunications our bank deposits and withdrawals would not be as easy or as fast as they now are. Other exciting applications include the digital library that will enable you to read any article or book without having to go to the library. 20) What Lies Ahead (Ch. It allows our entire medical record (in archives or observations taken in real time). e-mail and Information Services (Ch. 17) Distributed Multimedia Applications (Ch. Other applications. etc. 21) Figure 1. even air-mail. is the transfer of money by EFT. In Chapter 17 we mention e-mail. 18) Teleworkers. though we may want the ability for more self-control over the content. like home shopping.6 Applications of telecommunications like banks. Its usage will increase as the usage of computers in the home increases.’ Such money transfers are for billions of dollars a day all across the world. weather forecasts and education. Email including foreign mail through the Internet is often available through local information service providers accessed by the telephone. It may well become as ubiquitous as the telephone or TV. These issues are examined in Chapter 18. Again.Introduction Message Handling and Related Applications (Ch. CompuServe was one of the three largest on-line service providers with around two million subscribers. 30% of all homes in the US owned computers and computer sales surpassed TV in annual sales for the first time ever. These applications along with interactive games will change the way we entertain ourselves. Information services may well be at its takeoff stage approaching a killer application. films. ‘Money is merely information. Another message handling application is that of teleconferencing but with the coming of multimedia. there are problems of security. It is currently used extensively for correspondence (private and business) as well as for copying (downloading) computer programs residing at other computer server sites. including X-ray or CAT-scan pictures. These and related problems as well as their solutions are the subject of Chapter 17. These providers also offer many services that include entertainment. Applications still in the development stages include the delivery of video-on-demand. with over two million subscribers joining just one information provider (AOL). transactions may not be as safe either. Of course. Some of the services are interactive such as chat sessions where one can exchange views and information from someone that you may not know and someone who may be across the oceans. privacy and economics. 19) Internet and Cyberspace (Ch.

at least not yet. Currently businesses do a lot of their communications and some of their advertising on the Internet but not much business in the sense of sales. With telecommunications we have problems of interconnectivity plus integration that can eventually lead to computer applications across LAN/MAN/WAN/GAN Internet Information superhighway Open Systems/ platforms/protocols/ objects Analogue world Digital world Broad bandwidth (Gigabits) Narrow bandwidth Video communications Multimedia End-user friendly systems User unfriendly systems FUNCTIONAL APPLICATIONS WIRED CITIES TELEMATIQUE SOCIETY Figure 1. cybermoney. Without telecommunications we have problems (and solutions) of logical integration of files. The problems of security and privacy of information are among the issues to be examined in Chapter 20. There is much talk about cybercash. This is because there is not yet any safe way to transact money on the Internet. Discussing the Internet will allow us to enter some of the space of cyberspace.7 Trends from past present to present future 8 . teleworking is a viable and attractive alternative to the crowded downtown office that must often be reached after fighting traffic jams and traffic lights. Chapter 21 is on integration towards a global systems through telecommunications.Telecommunications and networks Will computers and information services become as ubiquitous as the telephone and the TV? Will they be as end-user friendly and accessible as are telephones and TVs today? Will it take two to three decades to be accepted in the mainstream as it did for the telephone and TV? Must information services be regulated? Will all this information around threaten our privacy? Some insights into the answers to such questions will be found in Chapter 19. If you cannot afford the monthly subscription of an information provider and do not have access to a LAN (through your employer or university) then you can always go to a caf´ like the Caf´ Cyberia in London e e where for an hourly payment you can surf the Internet. This is telecommuting. The Internet is the subject of Chapter 20 and has been mentioned earlier as an outgrowth of LAN/MAN/WAN ARPANET/Internet Proprietary systems/ platforms/protocols objects ARPANET and LANs as well as in the context of information providers. With the boundaries of the workplace getting ‘fuzzy’. Before we get to Chapter 20. where it is used not only by individuals but increasingly by businesses. Telecommuters are also big users of information services. we discuss one other application that depends on computers and telecommunications. These issues are the subject of Chapter 19. digicash and digimoney. which is working at home using a computer and being connected to the corporate database through telecommunications. but you are advised not to trust your credit card to cyberspace. Telecommuting will require special resources and raises many issues especially of productivity and evaluation. or in Chapter 20 which is on the Internet and cyberspace. especially of e-mail.

. and not just a city but a region and eventually anywhere in the world that we so desire. In the next few years we will see improved technologies and infrastructures. others have the connections into homes and offices. transform proprietary systems to open systems through standardization. This is difficult to predict because it requires predicting not only the technologies related to telecommunication and networking but also the environment where telecommunications and networks will be used. Much of the future of telecommunications will depend of the response of the end-user and consumer as well as on the computing and telecommunications industry. Each of these industries are a multibillion dollar industry just in the US alone. Possible combinations of firms in these industries are many and sometimes referred to as the metamedia industry.Introduction space and distance. 1993: p. the process will be exciting. ee Our final chapter also looks at the future from a historical perspective. There is a growth in the demand of information related services leading to an information-intensive global society that is propelled by consumer markets. Whatever the final product offered and whatever the delivery mode. All around the world new industrial structures are rising out of the deregulation of traditional PT&Ts (Post Telephone and Telegraph). One view is that there may not be any dramatic breakthrough in the technology of telecommunications in the immediate future but that we will continue to evolve on a steady growth curve consolidating much of what we have. software. . . These trends and evolution are summarized in Figure 1. . The cable company did not budge. as the French Norma and Mink called it. Although many of the information service providers are likely to emerge from the telecommunications and computer industries. . One entry into this new industry was announced in the US in early 1994 between the cable company TCI and the telephone company Pacific Atlantic. (Nazem. enhance LANs. . the information service industry will continue to evolve with the passage of time. and even the publishing and entertainment industries. cable. various valueadded information related services are opening up to competition. They were to have over 30 billion in assets and promised a 500 channel two-way interactive video-on-demand entertainment. enlarge the narrow band and single media to the broadband and multimedia. There may well be more failures. We can then integrate applications not only in a corporation but in a city. While the telecommunications industry will provide the network for the information intensive society and the computer industry will provide the advance processing capability. the t´l´matique society. In about 100 countries. Of course this requires an international infrastructure and international standards and a few other prerequisites that will enable us to reach for a wired world or. Competition and ease of entry will accelerate the creation of new services and new markets. We shall revisit this figure at the end of the book in Chapter 21 and then identify the technologies that led to each of the transformations. new providers will create their own niches in the market in the future . But then the regulated rates for TV were reduced and the telephone partner wanted a reduced price to buy. new and enhanced servers with marked differentiation and specialization for different applications. and so before the year was out the proposed merger was dissolved. . There will also be changes in related industries such as those of hardware. Thus we will convert the analogue world into the digital world. 19) The fierce competition ahead may well result in a better integration of delivery and services offered to the consumer but the shakedown may take long and will most likely be painful and costly. and increasing competition in the telecommunications and network industries. This will enable us to communicate and cooperate with each other in ways that were not hitherto possible and 9 . We are entering as era when global information networks and services will become reality. telephone. The new technology creates new demand. and foster the growth from functional applications to the wired and t´l´matique society through integraee tion and the interconnectivity of telecommunications. some have the technology and experience to integrate. but meanwhile there is much experimentation going on in the US and in Europe with different media and a varied mix of services to test the public on what they want and what they are willing to pay.7. Some companies have the cash to buy. WAN and the Internet. Eventually there will be a winner and many winners before the new industry stabilizes. .

2: Networking at the space centre The space centre at Houston. The new centre has over 19 kilometres of fibre optic cables connecting its hundreds of PCs and workstations with the larger computer systems owned by NASA. 198O The FCC in the US deregulates telecom equipment at customer premises and allows AT&T to offer tariffed data services and computer companies to offer non-tariffed communications services. Texas. The design specification is a commentary on the state-of-the-art of telecommunications and networking. thesis outlines the Ethernet. the firm started up its backup generator and was in full operation within the day. In addition. This damage included $300 million to the physical plant and infrastructure damage disrupting service to about 285 000 of the 1. 1972 SNA by IBM offers the first systems network architecture for a commercial network. 1974 Robert Metcalfe’s Harvard Ph. 1978 Xerox Company. One of them was an information provider that had planned for an earthquake in Kobe even though the odds of an earthquake there were very small. but in Singapore. In 1995. has controlled the flights of all the early spacecraft. Japan In 1995. survived because they had good network management and a plan for disaster recovery. the Nippon Telephone and Telegraph Company. Even a large and in some ways very important real-time system can be constructed from products that are commonly available and are no longer ‘high tech’. Supplement 1.44 million circuits in the region and knocking out over 50% of the overall services offered by the national telecommunications utility NTT. an earthquake struck Japan at Kobe killing more than 5000 people and causing damage of over $100 billion.D. a new command and control centre was partly operational and entered its beta phase of testing to replace the old centre and to prepare for the space shuttle into the twenty-first century. Some businesses. The problems still facing us are those of standards which is thwarting competition The high level of continuous technical communication in the telecommunications and computing industry has resulted in the industry refusing to settle down. It also had a backup generator to ensure that the system could be up and running even if all the local power lines had snapped. 10 Communications. it had a backup centre. Many businesses were severely disrupted and even the recovery operations were greatly hampered because of the lack of telecommunications. Intel and DEC give the first Ethernet specification. 47 8.1: Milestones for network development 1969 The US Department of Defense commissions ARPANET for networking among its research and academic advisers. This firm had leased lines from NTT serving its main offices in four cities in Japan in addition to leasing domestic satellite services from a VSAT (Very Small Aperture Terminal) satellite installation to bypass the domestic network. not in Japan. however.25 is the first public networking service. Source: Data pp. July. The next chapter is another summary and overview but only of the technologies to be examined in Chapters 3 8. 1974 Vinton Cerf and Bob Kahn detail the TCP for packet network intercommunications. Case 1. 1976 X.Telecommunications and networks could well result in the redefinition of the old paradigms of communication and work. When the earthquake hit. One design specification of this complex and important networking systems was that almost all the equipment must be ‘off the shelf’. 1995. Case 1. This provides the end-users with more choice but for the network manager it is a greater challenge. . This was specified in order to keep maintenance easy and not as costly as in the previous centre.1: Network disaster at Kobe. These computers and workstations are all interconnected in addition to being connected to the tracking stations all around the world.

Communications of the ACM. 21(9). and Rockart. Special issue on Survey of International Telecommunications. networks and the corporation. D. 3 19. D. A. and future. 25 28.) 1993 British Telecom buys 20% of MCI and this marks the beginning of a truly global market. 1 52. From LANs to GANs. Sankar. . 1985 The Japanese government enacts the Telecommunications Business Law.D.. The newboys: a survey of telecommunications. 33(2). Financial Times. 1982 Equatorial Communications Services buys two transponders and the Weststar IV satellites. present. pp. and Salameh. Scientific American. 19 July. C. Malone. D. 23(11). (1992). see Chapter 20. Weinstein. 5 Oct. Nazem. Telecommunications and the information society: a futuristic view. ISDN may be here to stay . 1984 AT&T divests ownership in local telecoms. L. S. 214 224. Telecommunications. Campbell-Smith.Introduction 1981 IBM introduces the personal computer. Carr. 26(7).A. 1989. IEEE Spectrum. 1990 The ARPANET is officially phased out and the Internet is born. Data Communications. But it’s not plug-and-play. Tillman. H. 28(10). It is licensed and a regulatory authority is established.W. Telecommunications in the coming decades. 128 136. 265(3). and Kiaster. Computers. and Yen. The spirit of networking: past. 62 67. SNA and OSI: three stages of interconnection.S. Economist. J. 11 .F. networks and work. 27 33. PC. Scientific American. Sproul. 6(1 & 2). 1984 The UK’s Telecommunications Act authorizes the privatization of British Telecommunications Ltd. (1987).B. . 265(3). T. J. Computers. 116 123.R. 1ff. (1991). W. 1987 The Commission of European Community publishes the Green Paper which calls for open competition in the supply of equipment and the provision of data and value-added service. (1991). (1991). Telecommunications. (1993). (1994). S. M. S. which abolishes the monopolies of the country’s domestic and international carriers. Information Management Bulletin. (Chi-Chung) (1990). (For milestones in the life of the Internet. Doll. Bibliography Budway. (1992). and Dent. 23 27. giving birth to the first very small aperture service (VSAT) industry.

This Page Intentionally Left Blank .


This Page Intentionally Left Blank .

Database services also fit into this category. With teleprocessing (the processing of data received from or sent to remote locations by way of a telecommunications line. You can see why the term ‘teleprocessing’ is often used as a synonym for telecommunications. Local area networks (LANs) which permit users in a single building (or complex of buildings) to communicate between terminals (often microcomputers). 2. According to the director. Networks may be: 1. The rise of distributed data processing When computers were first introduced. A combination of the above. such as coaxial cable or telephone wires). organizations did not take advantage of Grosch’s law (applicable to early computers) which states that the increase in computational power of a computer is the square of the increase in costs. 4. for example in a distant sales office. International (wide area) networks.2 TELEPROCESSING AND NETWORKS In 1899. that is. This chapter surveys the technology or telecommunications. the collection of data from many sources and the sharing of expensive computer resources. most organizations established small data processing centres in divisions needing information. advanced in the 1970s to allow the linkage of workstations. the most difficult to implement because standards and regulations governing telecommunications vary from country to country. input and output is instantaneous. Likewise. Linked LANs within a small geographic area.or picoseconds. particularly when users are not located in the same building as the CPU. 5. doubling computer costs quadruples computational power. peripherals and computers into networks. users will not get the full benefit of this speed if tapes and disks on which input is recorded are physically transported to the computer for data entry. everything that could be invented had been invented. branch office or warehouse. discusses the importance of telecommunications to business and looks at the problems of connectivity that corporate managers must resolve. This is the mode of processing for multiuser systems where people located in dispersed locations share a computer but need to input data and access up-to-date information at all times. the director of the US Patent Office urged President William McKinley to abolish his department. the centres were often poorly run. interact with a computer host (normally a mini or mainframe) or share peripherals. the most expensive networks because of long distances between nodes. 3. In addition. These centres were physically dispersed and had no centralized authority coordinating their activities. Networks are valued by organizations because they promote the exchange of information among computer users (many business activities require the skills of many people). the delivery of reams of paper output to the user can be time consuming. Because of the scarcity of qualified computer specialists. data communications and information communications. By failing to consolidate computer resources. Data processed in this manner were often slow to reach middle and senior management and frequently failed to provide the information needed for decision-making. National networks such as ARPANET to link computer users in locations across the country. they were unnecessarily expensive. which links input/output terminals to distant CPUs. nano. The technology of telecommunications. 15 . Although the processing speed of a CPU is measured in micro-.

Distributed data processing includes both the installation of stand-alone minis or mainframes under divisional or departmental jurisdiction and the placement of stand-alone microcomputers for personal use on the desktops of end-users. Users resented the red tape that computer centres required to justify and document requests for information services. from microcomputers to mainframes. feeling overworked. experience with data processing had given users confidence that they could manage and operate 16 their own processing systems without the aid (or intervention) of computer specialists. minicomputers with capabilities exceeding many former large computers were on the market at low cost. users no longer had physically to enter the centre to access the computer but could do so from a distant terminal connected by telecommunications.) Figure 2. Furthermore. Star Ring Star-Star = Host computer = Node computer Ring-Star Bus Figure 2. computer technology was advancing. A host computer may provide centralized control over processing as in the star network or the nodes may be coequals. (Failure of the central computer impairs processing for the entire system if the host computer breaks down in the star configuration. However. We now look at equipment configurations and technology to support teleprocessing and networks. Desktop microcomputers were for sale. computer specialists at the centres chaffed at criticism. By the time third-generation computers were installed in computer centres. faster delivery of output.Telecommunications and networks The need for centralized computing facilities was soon recognized. Chip technology had increased CPU and memory capacity while reducing computer size. (These nodes may be computers of all sizes. With centralized processing. underpaid and unfairly reproached by those with no understanding of the problems of systems development and the management of computing resources.1 Examples of DDp configurations . not all of the expectations for improved service were realized when centralization took place. The ring structure overcomes this problem because rerouting can take place should one processing centre or its link fail. elimination of redundancy in processing and files. Complaints about slow information delivery and the unresponsiveness of the centres to user information needs were received by corporate management. But it is generally associated with the linkage of two or more processing nodes within a single organization. Firms hoped that centralization would result in lower costs.1 shows sample DDP configurations. centralized computer. While consolidation of computing was taking place. increased security of resources and greater responsiveness to the information needs of users. This is discussed further later in this chapter. it was not. In turn. But generally networks contain a mix of equipment from different manufacturers which complicates information exchange. but could be linked to headquarters or to other processing centres (nodes) in a network. Computers were much easier to operate and maintain. tighter control over data processing. which facilitates linkage. Although DDP sounds like a return to the decentralization of the past. Strides in telecommunications meant that no processing centre had to be isolated. teleprocessing (also called remote processing) became the norm. each centre with facilities for program execution and data storage. This general dissatisfaction with operations led to a reorganization of processing once again to distributed data processing (DDP) the removal of computing power from one large centralized computer to dispersed sites where processing demand was generated.) The hardware at each node is sometimes purchased from the same vendor. Timesharing had also been developed whereby several users could share simultaneously the resources of a single. By the time DDP was initiated. large.

illustrates channel differences and lists applications for their use. A range of transmission options exists by combining different types of channel. telegraph. There is neither any indication of readiness to accept transmission nor any acknowledgement of transmission received. Some channels carry voice transmissions. Can be in both directions in sequence In both directions simultaneously A B Walkie-talkie Intercom A B Duplex or full duplex Picture-telephone Dedicated separate transmission lines (such as a presidential ‘hot-line’) Figure 2. transmission speeds and bandwidths the cheapest and most limited being a simplex telegraphic-grade channel. telecommunications lines that serve the public are licensed by the Federal Communications Commission (FCC). Public or private carrier? In the USA. television.2. facsimile and telephoto are sample communications channels which vary in the types of data or information they transmit and transmission features. line or channel enables communication of information in one direction only. the most expensive and versatile a fullduplex broadband system. There are over 2000 telecommunications carriers available to the public (called common carriers) such as AT&T for telephone and Western Union for wire Type Transmission direction One direction only Graphic representation A B Example Simplex Radio Television Halfduplex One direction only at any one time. When planning for telecommunications. These give a measure of the amount of data that can be transmitted in a unit of time. satellite. telephone. but this involves a delay when the direction is reversed.Teleprocessing and networks Transmission channels Data or information may be transmitted a few feet within a single office building or over thousands of miles. Types of channel A simplex communications. radio. In most communications lines. No interchange is possible. A half-duplex system allows sequential transmission of data in both directions. corporate management must consider what type of transmission channel is most appropriate for organizational needs and whether to use private or public carriers. 1 baud is 1 bit (binary digit 0 or 1) per second. cable.2 Types of channel in telecommunications 17 . An advantage in computer processing is that output can be displayed on a terminal while input is still being sent. Wire. The ability to transmit simultaneously in both directions requires a duplex or full-duplex channel. The capacity of the channel is measured in bandwidths or bands. Figure 2. a more costly system. some data. Current technology allows voice and data messages to be carried long distances over the same line at the same time and at an affordable cost. Transmission speed or signalling speed is measured in bits per second.

are a third option. switched services. An alternative to a common carrier transmission facility is a private data network. Furthermore. they will transmit data to and from equipment sold by many different manufacturers. A variation of the latter is a packet-switching service which breaks a data transmission into packets. Most packet-switching services have another advantage as well: they support a standard protocol (rules governing how two pieces of equipment communicate with one another) like X.3 for an illustration of Ethernet use. Some provide point-to-point service on a dedicated line. Connectivity and protocol support is provided for the equipment of many vendors. and transmits the packets over available open lines. (See Figure 2. Some LANs are vendor specific: that is. and Xerox’s Ethernet. That is. Private branch exchanges (PBXs).Telecommunications and networks and microwave radio communications. primarily telephone systems that connect hardware. Packetswitching networks can support simultaneous data transmission from thousands of users and computers. they support connectivity only between hardware manufactured by one manufacturer or manufacturers of compatible equipment. which explains why they are called local area networks (LANs). As a result. each containing a portion of the original data stream.) Some LANs are all-purpose networks. a user at a single terminal can access non-homogeneous hardware connected in the network. cabling and multivendor support. including speed. the data in the packets are reassembled in their original continuous format. nor does any corporate manager want a system that will quickly Other ethernets Office workshop Gateway∗ Printer Processor Ethernet cable Information processing centre Electronic file cabinet computer Typing system Printer Interface Production workshop Production machine Micro graphic cell Terminal in office Ethernet multiplexer Terminals Electric printer Figure 2. Such networks are economically feasible over short distances. capacity. A shared rather than dedicated point-to-point communications line reduces the outlay of a company for long-distance communications circuits. others. no company wants to invest in a system that will require the replacement of existing hardware or the addition of costly interfacing equipment. The network must fit into the existing environment and meet the organization’s functional needs. Choosing between these options is a difficult task that involves many technical issues.3 Ethernet 18 . Wang’s Wangnet.25 which is vendor independent. Upon arrival. sometimes in a roundabout route to reach a final destination. routing data through exchanges and switching facilities. Examples of such networks include IBM’s token ring network.

equipment is required to convert digital data to analogue signals (a process called modulation) when a message is sent and to reconvert the waveform Transmission 0 1 0 1 1 0 Modem Modem 0 1 0 1 1 0 Digital signal Analogue signal Digital signal Figure 2.5 Multiplexer 19 .4 Digital and analogue signals Terminals Multiplexer High speed full duplex Low speed half-duplex lines Figure 2. they rely on the Interface equipment To transfer information by telecommunications. many computer systems must add interface components: that is. To illustrate. most terminals produce digital signals (pulses representing binary digits).4).Teleprocessing and networks become obsolete or outgrown. hardware and software to coordinate the receipt and delivery of messages. A router not only retransmits but determines where messages should be forwarded. A peripheral called a modem performs this conversion. engaging terminals ready to transmit or receive data when channels are free or sending a busy signal.5). A bridge has a similar interface function but retransmits between two different LANs of homogeneous equipment. a name derived from modulation and demodulation. a multiplexer may be added to combine lines from terminals that have slow transmission speeds into one high-capacity line (see Figure 2. whereas many telecommunications lines transmit only analogue signals (transmission in a continuous waveform). a repeater acts like an amplifier and retransmits signals down the line. As a result. In addition. A gateway connects networks that use different equipment and protocols. For long-distance networks. Sometimes a number of terminals share a channel (or channels). A concentrator is equipment that regulates channel usage. back to pulses (demodulation) at the receiving end (see Figure 2. (Although managers should be familiar with these terms. The variety of LAN products adds to the dilemma and the intense competition among vendors to sell LAN systems puts pressure on corporate managers at the time a network decision is being made.

Keep teleprocessing statistics.Telecommunications and networks expertise of telecommunications specialists for network design. Perform message switching between terminals. As the management of each computer system may not be able to afford all the resources that they need. Large mainframe computer systems generally include a front-end processor programmed to relieve the CPU of communications tasks. 7. 4. file capacity and peripherals that include fast printers and optical scanners. 3. Facilitate the use of the CPU by several users in a time-sharing system. When Voic e gra de le vel Remote terminals Multiplexer Wideband channel Front-end programmable processor Host computer Keyboard terminals CRTs Touch-tone telephone Other terminals Figure 2. Provide access to external storage and other peripherals. Figure 2. Interconnectivity Each computer system may have a unique configuration of computing resources such as computer speed. Front-end processors may also: 1. Check security authorizations. a front-end processor may receive messages. the linking of computer systems by telecommunications and networks. store transmitted information and route input to the CPU according to pre-established priorities. It is telecommunications that provides the link and connectivity between computers that enables the sharing of resources and communication between users of different systems. Act as multiplexers and concentrators. 8. a disk server is a component that acts like an extra disk drive: it is usually partitioned so that each computer can access a particular private storage area. Another major function of frontend processors is to compensate for the relatively slow speed of transmission compared with the processing speed of the CPU. It may validate data and preprocess the data as well. For example. Process data when teleprocessing load is low or absent. it is desirable to be able to share resources when they are not being fully utilized. A file server is more sophisticated. allowing access to stored data by file name. This can be achieved through interconnectivity.6 Example of a teleprocessing system 20 . peripherals and interconnections with other networks may have a component that caters to all the requests of the networked computers.) A LAN of microcomputers. Accept messages from local lines with mixed modes of communication. 2. For example. 5.6 illustrates a sample teleprocessing system that incorporates some of the equipment described in this section. 6.

Multimedia will also be part of the discussion on information services and telecommuting. telephone traffic. The Internet is perhaps an area where the growth was much larger and faster than any one had predicted. metropolitan area network. The ‘bottom-line’ measure of the worth of telecommunications and networks in the 1990s will be in its applications. How will these integrated networks affect business communications? Information transmission will be faster. global area network. factors that traditionally favour the customers by leading to lower costs. For example. Salespeople may be able to sell and deliver their products without ever having to make personal calls to customers. E-mail and many applications in telecommunications in the early 1990s were textual. a subject discussed in detail in Chapter 17. we have a GAN. When equipment breaks down at a location. These applications are discussed in Chapter 18. However. The most popular design for a global network is called Integrated Services Digital Network (ISDN) initiated in 1984 by the International Telecommunications Union. medical services. and in the discussion of the Internet. A single network will often suffice. An engineer at a construction site will be able to look at electronic blueprints simultaneously with the architect at the office who drew up the plans. A GAN providing international connectivity is also known as the Internet and is discussed in Chapter 20. will be a high priority in the late 1990s. this group has not yet agreed on standards for hardware and software standards that are required if computing power. Some applications like EFT (Electronic Fund Transfer). And when the interconnectivity is global. images and data messages simultaneously over the same line at low cost. These communication services are part of message handling. each serving a different purpose (eg. not just for transfer of data but for selling and buying products with secure international transfer of funds on the Internet. A reporter covering the earthquake in Japan may send photos to London headquarters for distribution in an electronic newspaper delivered to the computer screen of subscribers. Many companies are currently part of several communications networks. entertainment and many a business where communications will no longer be by letter or even e-mail but by teleconferencing and video-conferencing. When the interconnectivity is within a metropolitan area. wide area network. subjects discussed in Chapter 19. The price of telecommunications services will drop. phone companies are installing digital computer switches and supplementing low capacity copper transmission lines with microwave and high capacity fibre-optic cables that will transmit information more than seven times faster than current rates. we have a MAN. The US government provided equipment and training facilities to do this. EDI (Electronic Data Interchange) and e-mail (electronic mail) will not change conceptually but they will be used more creatively and for a wider range of uses. Soon the stream of text will be integrated with other media such as voice and pictures giving us multimedia applications that could be very useful in education. Already the rewiring of Europe and the USA is under way to create a coast-to-coast network to carry voice. For example.Teleprocessing and networks interconnectivity creates a network at the local level we call it a LAN. Although the development and installation . revenues generated by telecommunications are high. information and telecommunications are to be integrated in a 21 Networks in the 1990s The 1980s was a decade in which a large number of LANs were installed. This is discussed in great detail in Chapter 5. Controlling the content and privacy on the Internet and improving global communications. facsimile machine). WAN and GAN are compared in Chapter 6. New services will become available such as cellular and mobile phones. many of these LANs will be joined into national and international networks. In the 1990s. costs of integrated networks are staggering. when extended to a wide area it is known as a WAN. LANs. a subject examined in Chapter 20. Telecommunications will become more reliable. transmission will be routed to avoid the bottleneck. The market is growing and competition in the telecommunications industry is fierce. an organization of the United Nations comprising telephone companies around the world. in 1996 for the first time the US troops abroad (in Bosnia) were communicating daily with their families at home by email. The MAN. local area network.

such as word processing and spreadsheets. the success of ISDN may accelerate sales. Is ease of applications development important? Many users who write their own programs find development tools for personal computers easier to use than shared-logic applications development tools. together. processing and information exchange at low cost. In choosing an appropriate system. 3. To illustrate. A computer network does facilitate data collection.) A manager must be familiar with the strengths and weaknesses of both shared-logic systems and LANs in order to evaluate their relative benefits and trade-offs. the incremental cost of adding resources to a LAN is low whereas expanding a shared-logic system may require a complete change in the CPU. 4. If for general use. called a sharedlogic system may be preferable to the installation of a local area network. a private data network run by IBM permits the exchange of information between companies with compatible IBM machines. If this network were meshed with ISDN. telecommunications add to the responsibilities of corporate management as explained next. How dispersed are users? LANs are not designed for wide area access. With a shared-logic system. In addition. looking for ways to attract telephone defectors to network services of their own.Telecommunications and networks single transportation system. (A shared-logic system utilizes terminals connected to a centralized 22 . technical and market tests for integrated digital networks have been made by the state-owned telephone companies in Germany and Japan. applications that must be integrated with Issues facing corporate management The quality of decision-making by managers should improve with integrated data networks because more information and more timely information will become available on which to make decisions. Most serve a single building. 8. 9. usually microcomputers. For computer manufacturers. Are concurrent requests for information from databases likely? Shared-logic systems are better able to respond to such requests. For instance. In fact. However. most provide file and record locking and offer transaction logging and recovery facilities. computer in which all processing occurs. Companies already in telecommunications and computer companies wanting a share of the telecommunications market are likewise following ISDN projects with interest. 7. the following questions should be asked:. Is data security an issue? A LAN allows decentralized data under user jurisdiction. perhaps a multiuser system is not necessary after all. a shared-logic system would be advisable. shared-logic technology should be favoured. then a LAN is appropriate. In addition. 1. Are gateways to other computer networks required? Although much work is currently being done on gateway technology. and the sharing and distribution of information within and between departments. Local area networks tie otherwise independent computers. (If files do not require constant updating by many different people. How are computing resources used in the company? If the system is primarily for high-volume transaction applications. 5. Are users willing and able to take responsibility for systems operations? If not. Is peripheral sharing a primary requirement? Sharing is convenient and cheap with a LAN. Nevertheless. cost-effective usage of computing resources. the array of options in the organization of computing resources is what makes the manager’s role so difficult. Organization of information resources The duty of corporate management is to plan for data access. the network would be able to expand its services.) 2. Is growth expected? LANs can be upgraded with ease since each added workstation brings its own CPU resources. but is not the only option. The success of ISDN will affect all companies with a vested interest in telecommunications. a multiuser system that uses time-sharing to link terminals. 6. Numerous ISDN trials conducted by American telephone companies are also under way. a security officer can impose strict control over data use and storage. according to some detractors) before differences can be resolved. It may take years (possibly decades.

Organization of telecommunications and networks is crucial to the orderly operations and growth of most computing. Network management is the subject of Chapter 13. capture. is to have industry standards for hardware manufacture such as the ISDN mentioned elsewhere. These organizational structures are not mutually exclusive. Security is also an issue with computer systems especially when they use telecommunications and networks for now they are exposed to many sources of infiltration and systems violation. there are tasks and issues that face network managers. a microcomputer can be hooked up to a mini or mainframe with excess capacity and used to create data. In addition. destruction or disclosure. and input/output inefficiencies are a concern. Will the network contain products of different manufacturers? Many LANs enable such connectivity whereas connectivity between products of different makes is minimal with most shared-logic systems. In today’s market. In general. The resources managed are examined in Chapter 14. at least on the hardware side. Data/knowledge must now be protected from unauthorized modification. and the lack of application availability when needed. This is the subject of Chapter 15. for meaningful communication in a worldwide market. the network manager has a spectrum of choices in both hardware and software and must decide how best they are used for working as individuals or in groups. user pressure for personal computers. upload data to mainframe storage or download data for microcomputer processing. dissatisfaction with the current operations is the driving force towards the establishment of LANs. Communicating across national borders is becoming increasingly important in our worldwide economy. a LAN is considered when users need to share data. One of these tasks is the acquisition of the necessary resources needed for telecommunications and networking. But in the real world. and is the subject of Chapter 16. the client server approach. The result is a mix of systems with dissimilar architectures and operating systems unable to exchange information without ‘patch-work’ and inefficient interfaces. Summary and conclusions In the 1980s. Or employees with compatible hardware might pass around disks holding files to be shared. Thus far the discussion concerns network management at the corporate level. The problem is further complicated in networks and telecommunications because. many business organizations installed local area networks to supplement their computer systems by having interconnectivity 23 . However. telecommunications have to be global and standards have to be not just nationally agreed upon but agreed upon internationally. It is management’s responsibility to decide how telecommunications can best serve the company’s long-term interests. high costs. Its organization structure is examined in Chapter 9 with a popular configuration. so they often represent different generations of technology with the problems of connectivity and compatibility not having been addressed. 10. For example. One solution for interconnectivity. Selecting most (if not all) of the resources from one vendor is tempting for it will eliminate problems of interfacing with vendors and the incompatibility of resources. However. corporations must communicate with other corporations and individuals within the country and need a national infrastructure. management wants better processing control. stand-alone microcomputers may be on the desktops of some workers. Whatever the organization structure. This problem is addressed in Chapter 12. excessive downtime. In a shared-logic minicomputer environment. Many corporations have developed systems one at a time. the need to integrate new systems exists.Teleprocessing and networks other large systems are better served at present by shared-logic systems. examined in Chapter 10. A single firm may have one or more LANs to supplement a shared-logic system. This subject is examined in Chapter 11. computer products are put together as and when the budget allows. In-house computers may also be linked to external computer resources. reaching agreement on standards is a slow and difficult process. Thus the organization of computing resources can be tailored to the unique operating environment of each firm. a LAN might prove the answer to poor response time. There are other resource configurations to consider still. In organizations with a proliferation of personal computers.

These are the subject of our next chapter. One lesson that has been drawn from this sad story is that tomorrow’s technology should not be installed today without adequate preparation and good risk assessment. To enable us to discuss applications of telecommunications and the management of telecommunications and networks needed for these applications.2 Case 2.9 43.1 24.4 16. and the wireless and cordless channels.8 13. It has also been argued that in this situation the risk was worth taking. But telecommunications also adds to management’s responsibilities in areas such as the organizations of information resources and making them operational and secure.Telecommunications and networks and the capability of sharing resources.3 59. Telecommunications provide managers with more information and more timely information than in the past which should improve decisionmaking. If there were not some managers who took calculated risks in computing and telecommunications. 1996. travel agents in town and airports around the world.5 17. One subsystem was designed to deliver baggage from the plane to the airport building even before the passengers were ready to claim their baggage. Source: Data Communications. Such management of computing resources are examined in Chapters 9 16.7 23. In this chapter we took an overview of some of the basic telecommunication technologies. radio. p. In the 1990s.1: Delays at the Denver Airport In 1995. 4B. we need to know more about the technology of telecommunications. p. in addition to maintaining telecommunications not only in the airport but with pilots in the air. including government agencies) will participate in regional and national networks of linked LANs. teleconferencing and access to online remote database services used in offices today.4 20.2 19.6 27. This subsystem delayed the opening of the airport and the contractor for the subsystem claimed that they were rushed and they needed more time to install the system to start with but were not given that time. 28. and International Herald Tribune.2 31.7 37. The airport had a sophisticated network that automated many subsystems at the airport 24 . The result will be faster.0 17. then we would not have many of the applications that we now have today. The future trend is towards integrated digital networks extending nationally and internationally.3 21.8 13. EFT (Electronic Fund Transfer). July 1994. The $300 million subsystem used ATM technology (to be discussed later in this book) along with 55 computers and was designed to handle 30 000 items of luggage daily. In this chapter we looked at the front-end and the back-end of a telecommunications system which are discussed in detail in Chapter 4.3 18. This chapter is in a sense an overview of this book with an introduction to some of the basic technology of telecommunications and networks. Supplement 2. 30. the new airport at Denver in the US opened after long delays and a cost of $5 billion. How much longer had they wanted? Around 16 months which happens to be about the time for which the opening was delayed. It was designed to be the state-of-the-art structure designed for air transportation well into the next century. cable.8 39. In between are the transmission channels such as the telephone. These applications are the subject of Chapter 17 20. Feb. The details will be the subject of the Chapters 3 8.1: Top telecommunications companies in 1994 Company NTT (Japan) AT&T (USA) Deutsche Telekom (Germany) France Telecom BT (UK) Telecom Italia (Italy) GTE (USA) Bell South (USA) Bell Atlantic (USA) MCI (USA) Revenue Mainlines (Million US$) (Millions) 68. more and more businesses (and non-profit organizations. more reliable telecommunication services for the business community far beyond the present day e-mail (electronic mail).

J. 23 (2). Derfler. Telecommunications. 4 11 October. Soon. 31 41. (1991) PC Magazine Guide to Connectivity. 19 July. F. D. Derfler. Telecommunications. Remote access: major developments in 1995. Jr. Flanagan. How Networks Work. (1991). Jr. Special issue on ‘Survey of International Telecommunications’. 1995. 2 48. (1995). 72 81. and Freed. Doll. D. (1992). Networks. and future. 62 71. 11.J. Scientific American. Interfaces. (1993). Dertouzes. Ziff-Davis Press. Special series on ‘Telecommunications in Europe’. 57 58. 25 . 1995. Scientific American. 21 (9). present. P.G. (1994). International Herald Tribune. The ten hottest technologies in telecom: a market research perspective. p. L. V. Special issue on ‘Telecommunications’. The spirit of networking: past. 25 28. computers and networks. 28 (1). Ziff-Davis Press. pp.R. 1989. Communications. 12. Oct.L. F. 265 (3). Bibliography Cerf. 29 (5).Teleprocessing and networks Source: International Herald Tribune. Data Communications. (1991). Financial Times. 1ff. 265 (3).M. M.

It is easy to install connectors to a coaxial cable but the connectors must be good since a bad connection can adversely affect the entire transmission system. Even for short distances. and fibre is also more reliable. Each of these media will be discussed in turn for their advantages. Samuel E. It comes in one of many forms: solid or stranded. the latter is more expensive but more reliable. The main problems with copper wiring are threefold: it has a low capacity.3 TRANSMISSION TECHNOLOGIES The new mobile workforce doesn’t so much need computer devices that communicate as they need communication devices that compute. However. . stronger than steel. then copper cables are inadequate. We will examine the evolution of this technology and its revolutionary implications for the way we may communicate in the future. As distances increase and as larger demands are made on the capacity (by volume as well as the nature of traffic. or coaxial. Shielded twisted wires are relatively expensive and difficult to work with. Even the glass fibre has limitations for distance and then we need microwave or satellite capability. it is slow and it is adequate for only a short distance. The most common (and oldest) is the copper wire and its variations. Bleeker Introduction In this chapter we will look at the transmission media needed for communication. such as video demanding greater capacity). Fibre transmits pulses of information in the form of laser emitted light waves. we need glass fibre optic cables. limitations and applications. It is not only fast in transporting data but it is effective for greater distances than is copper. The shielding is required to protect the conductor from outside electrical signals and reduces the radiation of interior signals. It is aptly described by Arnbak as a ‘(R)evolution’. the stranding and/or twisting of a pair of wires provides a shielding reducing the absorption and radiation of electrical energy. A more recent transmission media is the cordless and wire-less person-to-person communications. Thus even for the short distances as for internal wiring in an aircraft. text. and as the need for greater speed becomes relevant. A glass fibre is thinner than a human hair. They are also difficult to install. fibre is used because it can carry a mix of multimedia traffic that includes data. Another cable made of glass fibre is more appropriate. The conductor carrying the electrical pulse that represents a message itself can be solid or stranded or twisted. Glass fibre is made of silicon. fibre is used to carry voice and music. and 80 times lighter than a copper wire of the same transmission capacity. The distance for fibre links is more than 11 times the maximum distance for coaxial cable and 15 times the distance for some twisted wire systems. unshielded. for longer distances and for a variety of traffic such as voice and video. a substance as common as sand. images and voice. a shield could be of woven of copper braid or metallic foil which 26 has the same axis as the central conductor and hence is referred to as a coaxial cable. Such connectors are often made of tin or silver. Such wiring is best for short distances and small capacities. Fibre is also reliable because it does not pick up extraneous electrical impulses and signals. Wiring The oldest and still commonly used transmission media is the copper wire. shielded. The capacity of a fibre is one billion times the capacity (in bits/second) of a copper wire (for the same cross-section). Whether single or stranded.

More on infrastructure and more on a wired city later. Statistics for a sample of geographic areas is shown in Table 3.4 Personal computers/100 people 28.1. where the transmissions for long distances requiring large capacities and carrying Table 3.1 Source: Stix (1993: p.8 42. But this advantage can also apply to developed countries like France that had an outdated telephone system in Paris.3 7.4 9. The different modes of wiring having different applications result in a coexistence of all or most of these forms of transmission in many a telecommunications environment.3. From the above discussion.2 1. This explains why it is cheaper (in total cost terms) to install an advanced technology starting from scratch (without an infrastructure) than replacing an old infrastructure. The density of these applications vary with countries. Their use would depend on the carriers responsible for the transmission and will vary with countries depending on their applications. This adds to the security of the systems and is very important when confidential messages are being transported.1 Comparison of wiring approaches Unshielded twisted wire Speed and throughput: Average cost/node: Media size: Maximum length: Difficulty in installation: Protection from electrical interference: Fast enough Least expensive Small Short Difficult No protection Shielded twisted wire Very fast Expensive Large Short Difficult Some protection Coaxial cable Very fast Inexpensive Medium Medium Moderate difficulty Good protection Fibre optics Fastest Most expensive Tiny Long Relatively easy Very good protection These signals are picked up by copper which becomes an antenna and absorbs energy from radio transmitters. and two.6 TV households with cable (%) 55. One is that replacing copper wire represents a loss of investment to the carrier owning the copper wire who should then be expected to oppose fibre in order to support his investment.9 Cellular phones /100 people 2. In contrast. Fibre and advanced switches were installed and a free computer terminal was given to every household with a telephone.1. power lines and electrical devices.6 27 . glass fibre cables are immune to electrical fields and so they do carry clean signals that never spark or arc and add to the reliability of fibre as a conductor.2 14. The characteristics of video compared to those of telephone. The average cost in the US of wiring a home with fibre is roughly $1500 compared to $1000 with copper wire. A recent application of telecommunications is the transmission of video.2 1.2 13.2.2 Services in selected parts of the world US Telephone lines/1000 people 48. cable TV or PCs (personal computers). The light waves in glass fibre can be precisely controlled and is less vulnerable to unauthorized access compared to the electrical pulses in a copper cable. A comparison of wired technologies for transmission is summarized in Table 3. This explains the advantage that developing countries without any infrastructure have. Paris today has one of the most advanced telephone systems and a infrastructure basic to a wired city. Note that the increment is only $500 per home but the total cost of having copper and then replacing it with fibre costs $2500. We must get back to transmission technologies. Also copper develops voltage potentials to the electrical ground resulting in interference. the laying of fibre in new homes represents a savings of $1000 per home over replacing the copper wire (and including the sunk cost). 104) Japan Europe 42. be they telephone. As a consequence. There are two observations worth making. one can conclude that the different media of transmission do have their distinct advantages and limitations.Transmission technologies Table 3. Glass fibre is much lighter and smaller than a copper wire but it is much more expensive. instead of a telephone directory (which at the time of the initial planning cost just as much as a terminal). One is shown in Figure 3. cable and PCs are shown in Table 3.

Earthbound or terrestrial microwave systems use high-frequency electromagnetic waves to transmit through air or space to microwave receiving stations. When a satellite is placed in geosynchronous orbit in space. Microwave A problem with wiring is that it is limited in distances for which it can transmit effectively. microwave transmission is necessary. Thus the microwave alternative is not ideal. so that it is stationary in relation to any point on . however. though it was a great innovation when it was first invented.Telecommunications and networks Table 3. These stations. It is now a reasonable economical option only for distances up to around 20 miles offering signalling speeds of 1.54 megabits/second. hills or other tall obstructions do not interrupt this line of sight. say for the connections to offices. For local connections for local TV. roughly 30 miles apart (on land) and the 28 transmission towers are tall enough so that the curve of the Earth. schools and businesses. coaxial cable is used. we use coaxial cable or copper cable.5 1 Entertainment Co-axial cable 1971-1988 Long Distance Telephone Signal converter for Cable TV Office School ROUTER Business Signal con- Video Router for Telephone & TV KEY Fibre Coaxial cable Copper cable Figure 3. Satellite A satellite in space can overcome the ‘line of sight’ problem besides being less sensitive to the weather. They are placed close enough to each other.1 Use of different transmission media a mix of traffic will use fibre. must be within ‘line of sight’ of each other. Weather conditions like humidity and dust can also affect transmission. Beyond that distance. its speed allows it to match exactly the speed of Earth’s rotation.3 Comparison of services in the US Telephone Cost as % income per capita: (Drop over the years) Usage: Transmission media: Period for adoption in home: Cable PC Video 14 2 Communication Copper wire 1876-1950 4 1 Entertainment Broadcasting 1950-1991 11 2 Computing Fibre 1975-1995 3. and for connections to PCs and telephones in homes.

Eliminating this wire gives us the cordless system.2. Wire-less/cordless systems Whether by microwave or by satellite. For a wider range of distances. or by other means of transmission. the longer the distance. One configuration of satellite transmission using Earth stations between an office and a remote factory is shown in Figure 3. These are connected through a wall-mounted transceiver (which connects thin and thick cables like the fibre and coaxial) to a wired LAN and establishes radio contact with other portable networked devices within the broadcast range. Even the name ‘wireless’ is as old as the Marconi’s Wireless Telegraph system of 1900. Microwaves are then sent to the satellite some 22 500 miles away and then sent back to Earth stations anywhere on Earth. The great advantage of the wire-less system is that it no longer is fixed and restricted to a physical space. within a building or home. Thus. you may use a wired network with laptops and palm top computers. You can now ‘roam’ around (roamers are subscribers in locations remote from their home-service area) anywhere you wish and still communicate with a phone that has a transmitter. Each satellite can transmit and receive signals to slightly less than half the Earth’s surface. it is an old technology since it really is still basically microwave broadcasting. For wider ranges. which may not be desirable for real time applications but good enough for many applications. the new system is sometimes referred to as wire-less or cordless. It is considered a new technology because it is recent in time but 50-64 million bits/second (per channel) Satellite Earth station Earth station Local station Local station 1-1 million bits/second Office Factory Figure 3. especially for day-to-day private and business data communications and even voice communications by telephone. Of course. the different the technology employed.Transmission technologies Earth. small antennas on the back of a PC or within PCs communicate with radio towers in the surrounding area and are referred to as cellular phones. To avoid the confusion.2 One configuration of a satellite communication 29 . therefore at least three satellites are required to effectively cover all the Earth. there always is a telephone wire at some end. The return journey by satellite could take up to half a second.

The one-way (unidirectional) may be either radio or message service like paging or downloading of data from either a local or a wide area network. a laptop. . Cellular technology had its birth in the 1970s in the labs of Bell in the US. the packet radio and the PCS. the CT-2. With developments leading to the second generation cordless technology. but instead we shall examine three popular systems: the cellular radio.3. the wireless LAN. Besides distance. The base-station can serve tens of wire-less terminals. digital and analogue. the cordless system has a wide spectrum of other combinations like local or long distance. and the CDPD (Cellular Digital Packet Data). or to modems in host computers which may be minis or mainframes or other PCs. Network control remains distributed. a European standard GSM (Groupe Special Mobile) was adopted by 32 countries. and time division with twelve channels per carrier (DECT). One configuration of the cellular phone is shown in Figure 3. (Goodman. satellites are used to pick up low powered signals from mobile networked devices. we see a change from stand-alone consumer items to elements of a geographically dispersed network. The second generation networks. There are other standards in the US and Japan.3 Taxonomy of wire-less/cordless transmission 30 . and two-way or one-way. This classification is not mutually exclusive. cable to devices like telephones or other cellular phones. Some have choices of their own. which transmits ‘packets’ of data through sparsely used radio channels or during gaps in conversation.Telecommunications and networks even around the world. The two-way (bidirectional) allows for a dialogue and its manifestations are either the cordless telephone. as well as the British standard. 1991: pp. Thus the digital computer could be either a palm-top. with wireless terminals competing with little or no central coordination for available channels . the Cordless Telephone Version 2. the portable computer. will employ a variety of access technologies including narrowband time division with eight channels per carrier (GSM). The PSTN is connected by fibre or coaxial Private networks Local (cellular) Public switching Packet switching Packet radio A packet radio is an approach that uses a switching and PAD (Packet Assembler and Disassembler) where the message being assembled Palm-top Lap-top PDA Two-way Cellular radio (PCS) Portable computer Cordless telephones Digital Portable WIRELESS Packet Radio Personal Long distance (satellite) One-way Data Transfer Messaging • paging • down-loading Analogue Figure 3. We shall not discuss each in detail.4. or a PDA (Personal Digital Assistant). . 33 4). whilst the cellular switch may serve up to 100 base-stations. portable and personal (PCS). all using digital speech transmission. Cellular radio A cellular radio transmits radio signals to a base-station to a cellular switch through broadcasting that connects the wire-less network with the fixed network by a cellular switch with the PTSN (Public Switched Telephone Network). frequency division (CT-2). or the cellular radio PCS (Personal Computer System). All these types are shown in a taxonomy of cordless systems in Figure 3. In 1993. narrowband time division with three channels per carrier (IS-54).

It is based on a digital architecture. Such a configuration is shown in Figure 3..Data ports can be inexpensively built into PCS handsets to allow direct transmission by bypassing the handset’s voice coder.Transmission technologies Air wave link Cellular data interface/ adaptor Fibre Cellular switch Fibre Mobile computer PSTN Fibre Host modem Copper wire Internal modem Copper wire Fibre or coaxial cable Computer Figure 3. (Wimmer and Jones.5. but the infrastructure for a PCS was well planned. Because PCS cell contours are relatively small. In the mid-1990s. there was much debate and argument about the assignment of frequency spectrum which hampered the development of the PCS. especially the UK. The base-stations not being addressed ignore the broadcast while the base-station that is being addressed accepts the message and passes it on to the ultimate destination. which are utilized by grids of low-power base-stations with relatively small cell contours. The packets are then relayed by the PAD through a satellite to the possible destination by broadcasting to different base-stations. 1993: p. which are self-contained sets of information. A defining technical characteristic of PCS is its high capacity and spectral efficiency. a Low Orbit Satellite System. which may be a device or host computer. In the US. 22). Assigned spectrum is divided into discrete channels. thus permitting PCS handsets to act as wireless modems for portable computer and facsimile machines. the PCS was still considered experimental in the US though it was well advanced in Europe. light and inexpensive. It was projected that 31 . PCS (Personal Computer Systems) A PCS is a LEOS.4 Cellular phone connection to corporate computer (or disassembled) into packets. PCS also will be useful for private in-building or campus-based wireless PBX systems because of the potential to assign frequencies to relatively discrete areas. PCS handsets can operate at low power and still be small..

2 gigahertz versus 800 megahertz. (Flana- Comparison of transmission systems A comparison of the two cordless systems (the packet radio and the cellular phone) with two other popular cord systems (the microwave and the mobile satellite system) is shown in Table 3. PCS requires five times more towers to provide the same service as existing cellular phones .5 Packet radio satellite broadcast the cellular industry will spend $20 billion on PCS infrastructure alone. Because of the high frequencies. The lure: lower mobile phone bills. Once the new spectrum becomes operational. New spectrum to play with and insurance that Table 3.g.4 Comparison of transmission approaches Applications Microwave ‘‘Long’’ distance transmission Mobile and long distance transmission Telephone Circuit Switching Roaming modem Packet switching Roaming location Advantages Easy installation Wide band Good compatibility Saturation coverage Reliable Voice/data supplier Mobility Matured technology Existing infra-structure Mobility User pays only for packets transmitted Disadvantages Poor penetration (e.Telecommunications and networks ! Message switching (PDA) ! Packet radio KEY: ! = Target Figure 3. but a way for large cellular operators to expand their networks. 1995: p. gan. PCS is not a separate or competing network with cellular.4. . PCS will deliver a technological advance that will draw multiple converts. . of walls and floors) Poor mobility Expensive Poor data rates Limited capability Expensive Poor penetration Not universally available No voice support Coverage limited to large metropolitan areas they will have bandwidth to expand. 32). Mobile satellite Cellular Packet radio 32 .

in a car (as a passenger of course). Mobility and portability may create a new class of applications that combine Cable Microwave transmission Satellite Cordless/wire-less Figure 3. and the cordless phone. It could be a tool for professions like the a salesperson or a medical professional. Transmission and telecommunications capability is fast becoming a tool for the office and part of our office culture. or even while walking. there have to be PC applications that know how to better use wireless data capabilities. There are a number of applications of wire-less computing that include POS (Point-Of-Sale). As identified by analyst Jay Batson. dispatching of repairmen or taxis. partly because they have all our worktools together and partly because the workplace creates an atmosphere conducive to work. they may never replace the transmission done from a fixed site like a work-site whether it be at home or the office. these technologies may well coexist for they all have distinct functions to perform.6 Micromotive transmission waves 33 . Summary and conclusions The main media of transmission are the wires that string our telephone poles. (Flanagan. the cables that connect long distances including below the oceans. to some extent. or pagers. To complement office computing. they are: Number one is a critical mass in a specific area so that the entire metropolitan area is covered. but also voice and even video. the microwaves that connect longer distances. there may well be handsets that deliver not just data. computing and personal electronics. There hasn’t been a kind of consciousness raised all the way up to the software level that people use on PCs. 1995: p. Each of the main transmission technologies has variations. These transmission technologies are shown in Figure 3. or a medical emergency while ‘roaming’. the satellite that enables much longer distances to be traversed quickly and efficiently. It enables us to perform when we think of it or when we need to make a transaction and not have to wait till we get to our office. And it can be ‘life-saving’ for a rural resident or in an emergency like a car breakdown. and even thereafter. etc. Telephone or telegraph line Wire-less technology does increase productivity by using productively the time we idle away whether this be at a waiting room. the cordless and wire-less transmission is in fashion and even a status symbol. What is needed to bring it all together into a useful and productive technology are two things.Transmission technologies These four systems may survive the shakedown of the industry as it matures. The symbiosis of computing and communications has applications and implications that we have yet to discover and appreciate. Despite the many applications of the cordless and wire-less devices. 38). delivery of services varying from pizzas to medical help. Soon it may be difficult to distinguish between computers with telecommunications capability and telecommunications devices with computing capability. Number two. Meanwhile. The workplace is ingrained in many cultures and their work ethic.6. However. The secret is for a compelling market use to evolve.

It is the permutation and combination of these variants that is most suitable for any one environment of telecommunications. Looking at the future. sometimes the technical person but increasingly the corporate manager. it is quite possible that we may see personal cordless communicators which will enable us to communicate with each other in an office building (no matter how tall or large it may be) and with people outside the building (wherever they may be in the world).8 A view of transmission in the future 34 .7. One such scenario is shown in Figure 3. Unshielded twisted Copper Shielded twisted Wired (site -based) Transmission media Broadcast based Two-way Satellite Portable computer (PDA) Coaxial Fibre Message switching Packet radio Microwave One-way data transfer Cellular radio (PCS) Cordless telephone Figure 3.7 A Taxonomy of transmission media KEY: Broadcast `Roamer′ person anywhere in the world with an antenna Figure 3.3. a different view of Figure 3. To make such decisions it is necessary to understand some of the technical aspects of these technologies. which is the purpose of this chapter.Telecommunications and networks A taxonomy of the main important variants are shown in Figure 3.8. Selecting the right permutation and combination is the function and responsibility of the manager.

2: CT-2 and PCs in the UK The UK has been a world leader in the adoption of the CT-2 (Cordless Telephone Version 2) and the allocation of spectrums to the PCS (Personal Communication Systems).9 Varying penetrations by transmission media The constraint to current cordless communication seems to be the limits of penetration through opaque objects like walls and floors in a building. a US telecom equipment manufacturer. which is the use of multiple frequencies to achieve reliable communication with relatively weak signals. a Taiwan equipment manufacturer. Muiridi Investments of Venezuela. a consortium of 18 Japanese’s companies. the Russian manufacturer of the Proton rocket. the Chinese manufacturer of the Long March rocket. Raytheon. and reinforced concrete steel (walls or floors in a building) limits spread spectrum.4 billion. Case 3. The total cost is expected to be around $3. a Saudi investment group. and Pacific Electric and Cable. a Thai cellular and paging operator. the Italian cellular and radio carrier. It has two bands. Nippon Iridium.. What is needed is greater access to such equipment and its integration with other applications. STET. In this chapter we have examined transmission technologies. Lockheed. In the next chapter we will examine switching and enabling technologies of telecommunications. Inc. China Great Wall. a US launch provider. Other owners in this private enterprise adventure are: Bell Canada Enterprises.Transmission technologies Reinforced concrete steel (walls/floors) Radio 800−900 MHz Infrared <10 000 MHz Microwave ≥10 000 MHz Cloth Sheet rock Concrete Spread spectrum (varying frequencies) Figure 3. Mawarid Group. These penetration limits are shown in Figure 3. concrete but not sheet rock limits microwaves. 35 . One of these we have mentioned frequently: switches.. Case 3. The current restrictions are not a problem for most of us and the current technology seems quite adequate. Krunichev Enterprise.9. It has been financed in 1993 by a first round of equity offering of $800 million of which Motorola purchased a minority interest. Cloth limits the penetration of radio and infrared. United Communications Industry Co. But there are other related and enabling technologies for telecommunications that are needed. The Iridium satellites are designed for LowEarth Orbit (LEO) using the TDMA (TimeDivision Multiple Access) technology.1: The Iridium project The Iridium project is a global satellite communications systems designed by Motorola (US) that allows customers to call and be called at any time and from any place using hand-held telephones directly through a constellation of 66 LEO (Low Earth Orbit) satellites in six orbital planes and in an orbit of 780 kilometers. The Iridium project is expected to be operational in 1998. the L-Band that is from the satellite to earth and the Ka-Band that is an intersatellite link with gateways and feeder link connections.

’ One potential drawback in all the three approaches is that the transmission will most likely be limited to relatively short distances of around 50 miles. In early 1996. 1996. Dunphy. (1992). (1995). 22 7). Derfler.5 gigabits a second. IEEE Communications. .4 GHz. higher frequencies. Flanagan. In relative terms. . three research teams announced that they has independently achieved the Holy Grail of high speed transmission: the terabit threshold of transmission at a trillion bits per second. The European research program for advanced mobile systems.Telecommunications and networks The initial decision of assigning market share to individual CT-2 licensees was too small to provide wide coverage with high quality and low cost . And so the industry set an informal goal of increasing transmission speeds to a rate of a trillion bits per second through an optical fibre. 29(4). Arnbak. Myths and facts about optical fibre technology. 30(2). IEEE Personal Communications. or conveying 12 million telephone conversations simultaneously’. . J. The European (r)evolution of wireless digital networks. With telecommunication. Personal communications services: the long road ahead.4 billion bps. 1 and 4) Bibliography Amedesi. or 2. . (1997). D. 47 48. pp. B. P. systems were input/output bound because the speed of the CPU computing was much faster than the speed at which data could be fed as input and results could come out as output. 2(2). The three teams used different approaches but all had one thing in common: ‘instead of sending one stream of light through the fibre. Cox. Telecommunications. Sky links. in which subscribers had to look for signs to determine where they could place a call.S.C. .. 1992: p. The date of the breakthrough seemed feasible for around the end of this century. 22) The British Department of Trade and Industry made a study of the spectrum from 470 MHz to 3. Source: International Herald Tribune. (1995). or are replaced by alternate means of communication. Subscribers typically find this inconvenient and demand more ubiquitous coverage . 74 82. and found that personal and mobile communications will increasingly displace fixed radio relay services in the 1 3 GHz region as the latter move to lesscongested. J. (Wimmer and Jones.J. Wireless communications: what is it? IEEE Personal Communications. The developments of high speed transmission were done by Fujitsu Ltd. they sent multiple streams of light. thereby multiplying the amount of information that can be transmitted.3: Transmission at a trillion bits per second All through the first four decades of computing. 12 17. Satellite communications in the year 2000: the view from Europe. Source: Wimmer and Jones (1992: pp. PC Magazine. Da Silva. 16(1). 33 40. 20 35. 2/3 March. almost five years before the deadline. 24). The advent of a new telecommunications service in the United Kingdom encouraged greater price competition and the creation of valued new services by existing market participants. the Nippon Telegraph and Telephone Company and a team from the AT&T Research Labs. This was the equivalent of transmission 36 ‘the contents of 30 years’ worth of daily newspapers in a single second. (Wimmer and Jones. The time for commercializing these high speeds is estimated at five years. (1993). Telecommunications. This is because light signals degrade with distance and have to be amplified and reconstituted en route. Case 3. 2(1). P. the bottleneck remained because transmission speeds were also much slower than CPU speeds. 1992: p. each at a slightly different wavelength. F. 23 28. Trillion bps capacities are necessary for carrying video and multimedia signals for applications like films and videoon-demand. United Kingdom licensees implemented a telepoint service. (1996). Telecommunications. P. and Fernandes. This is very expensive for multiple light streams.E. 31(9). as is done today.C. consider the speeds achieved as being some 400 times faster than the commercial systems in use today which carry 2. NE1 NE12. The bottle neck was likely to get worse since CPU speeds were still on the increase. 26(3). (1995). at a conference on Optical Fibre Communications.

PC World.T. Scientific American. Domesticating cyberspace. Imielinski. 225 244. 12(2). 359 382. Stix. Data Communications. Wireless laser networking. 37(10). 37 39. Wireless LANs. Lauriston. S. J. B. 22 27. Wireless computing. R. Wimmer.M. 42 58. 22(11). 23(3). 57 62. Connecting over the airways. (1993). Telecommunications. IEEE Communications Magazine. Data Communications. Global development of PCS. Jutlia. 10(9). 18 28. and Badrinath.Transmission technologies Goodman. Telecommunications. S. 101 110. (1993). A. (1992). Laurent. (1997). Rehab for Copper Wire. B. 29(6).J. An intersatellite laser communications system. PC Magazine. B. Steinke. IEEE Communications Magazine. 269(2). and Jones. Trends in cellular and cordless communications. (1993). Communications of the ACM. 12(14). 30(2). 31 40. J. (1994). Gunn. 26(7). (1992). D. (1992). 37 . (1991). T. K. Johnson. G.A. (1994). 55 59. Saunders. (1996). 29(6).R. LAN. The cabling cost curve turns toward fibre. 90 92. Wireless data: welcome to the enterprise.

You would pack your car tightly and get rid of all the redundant space in a truck or large car. the bridge. And. nor masters slaves. obeying or anticipating the will of others . if the shuttle could weave. without a hand to guide them. Aristotle Introduction In the previous chapter we examined various media for transmission. Suppose that you had to pay toll according to volume and you had a compact car worth of passengers and luggage.) In networks and telecommunications there will be no loss of life but probably a loss or at least distortion of messages being transmitted. you must follow the rules of the road. (Many Swedes and Finns lost their lives before they agreed to drive on the same side of the road. you need an address of the destination and detailed instructions on how to get from the origin to the destination. It is these components and elements of a telecommunications systems which we will examine in this chapter. Let’s go back to the car transport analogy. and gateway are the first topics to be examined followed by addressing and protocols like the TCP/IP protocol commonly used even by the ubiquitous Internet (over 25 million users worldwide). If you are travelling a distance that involves many changes and switches that you must make on. . This address cannot be interpreted by humans as in road transportation but by a computer and so it must be in machine readable form and be complete as well as specific without any danger of misunderstanding or ambiguity. The messages must of course have an address. You must pay by volume (kilobytes) of message transmitted and so it behoves you to pack your messages tightly. The same choice occurs with telecommunications. We conclude with a short discussion of these enabling technologies being smart and intelligent. chief workmen would not need servants. Can you imagine what would happen if an American driving on the right and an Englishman driving on the left were to share the same lane of a road and go in opposite directions? There would certainly be a loss of life. The router. There is much information and enabling technology that is needed. For transporting information. . there are rules of telecommunication called protocols that must be followed meticulously. a motorway (or autobahn or freeway) you must know about all the ramps to get on and off and know how to interpret the many signs and instructions that you will meet. say.4 SWITCHING AND RELATED TECHNOLOGIES If every instrument could accomplish its own work. Also. ATM which is used not only for data and text but also images and video will be mentioned but deferred to a later chapter. This is known as compression and is important not only for reducing costs but also for reducing the time of transmission. and the pick touch the lyre. There is one more important concept involved. of course. This chapter is somewhat of an overview introducing many basic . The address is interpreted by a router which includes 38 software that determines the best path (or only feasible path) between origin and destination and may involve many switches that direct traffic between networks and LANs. Would you take a truck or even a large car and pay for the empty space? Of course not. Compression is often done by a router which along with a bridge and a gateway are facilities that connect networks. But having a transmission media (and a computer) is not sufficient any more than having a car and a road will get you to where you want to go.

if the message is too large. because it may be slow. repeater and gateway A router is primarily an interconnection to a network. Sometimes the packet has to be fragmented so that the packet for one network is not too large for another. transit delay. Is route 2 the best because it is the most direct? Maybe not. It is a portal device. What is best will be a function of variables like cost. one may consider the simplest portal device: the repeater. it is merely a short-hand used in the trade) than a router but has a less efficient routing and flow control. it may be broken up into packets.Switching and related technologies concepts that will be examined later in the context of their interrelationships. etc. 1989: pp. However. and that was long before the ubiquitous Internet (Tanenbaum. and then to repackage and retransmit the message. has a direct by route 2. and has one intermediary node C in route 3. One important function of the router is to read (and follow) the instructions in the header (at the front of the packet or frame) that concern the router. to provide addressing for the next destination. which is like a small box which you can hold in your hand and connect the two segments of a network cable. Consider the simple case shown in Figure 4. So all possible routes have to be looked at and the ‘best’ for the assigned objective function chosen. then the router will find a ‘best’ path for the message. For a lower overhead. undetected errors. that event each packet can be sent on a different route and reassembled at the other end.1 Routing 39 . In D E Message A A Route 2 B Destination Origin C Route 3 Figure 4. costly. that is not misspelled. error prone. Besides.1. There are many routing algorithms. Tanenbaum discusses eight. where a portal is the meeting point between local and long distance services. The bridge exercises great discrimination over passing traffic by refusing passage to the other side of the bridge if the destination is not on the other side of the bridge. You then need another network portal connection called the bridge. The path from origin A to destination B has two intermediary nodes D and E in route 1. This would mean that each packet must be labelled to identify its entity and destination. bridge. A bridge has a higher throughput than a repeater and even higher thruput (no. to strip the data that is no longer necessary. it may be busy and congested and hence not a feasible path anyway. The message may be composed of frames which are logical entities. even a string of repeaters cannot extend a LAN beyond a few thousand yards. given the constraints such as cost and availability. mishaps in transmission. The router should also be able to assign priority to one message over another (maybe a message to a hospital for a dying patient!).) If there is a message that is initiated at the router. and there are many possible paths for that message. or. (The equivalent in business is a mail-room or a shipping dock. size and priority assignment. 197 203). The repeater retimes and regenerates digital signals before forwarding them. This reduces non-essential traffic. For performing its many functions a router has a high overhead. However. a bridge may also not guarantee the delivery of the frames Route 1 Router.

Because of the different protocols. the cells with values are those in the left hand triangle while the right hand triangle is empty. X XX XXX XXXX XXXXX XXXXXX XXXXXXX XXXXXXXX Half the matrix is empty and therefore a candidate for compression Figure 4. called MPR.Telecommunications and networks nor its mishaps. For example. But there are multiple protocol routers. 3. then we need a gateway. Without the appearance of advanced hardware technology like high speed microprocessors. But what is compression? Compression We discussed compression implicitly when we discussed stripping redundant data by the router. Like the router (and the repeater). the date of its foundation. The compression solution to this matrix is to eliminate the empty portion and thereby greatly reduce the message to be transmitted. Compression is also used in DSSs (Decision Support Systems) where we have a matrix to process. it is desirable that the message be as short as 40 possible and may need compression. it is now possible to provide software support for routing a number of highlevel protocols simultaneously and within a single system. We do this in road transportation in a city where we cannot broaden the road and so we couple two parallel roads and make each go one way only.2 Matrix for compression . This is what compression is about. if a frame exists beyond the maximum transit time allowed. Instead. etc. In transactional processing we do not always process an entire file with all its fixed information. which is ‘simply a router that supports more than one protocol. But there are limitations to increasing bandwidth and so instead we reduce the flow traffic. 2. The protocol must be converted to replace the control information from one network with control information required to perform the equivalent functions in another network.’ The MPR can be combined with a bridge to give a bridge router system. If the protocols are different. we process only the variable data in addition to an identification which appears as a header to the set of records to be processed. 1991: p. the bridge does check for the frame size allowed by the network to be used and allows for priority. 32). Compression by stripping is one approach to increasing throughput through a transmission media. Bridge/routers work by routing the routable protocols and bridging the non-routable ones. the frame may have to be discarded. Sometimes this matrix has a ordered set of cells that are empty. This is important in wide-area networks and will be discussed later in Chapter 6.2. This can be a complex problem if the protocols at each level of the systems architecture have to be converted. a gateway must perform three distinct functions: 1. like the router. then a congestion of links and the bridge may occur with frames being lost unintentionally. Whatever network linking facility is used. a bridge is protocol independent. Compression is not new to data processing.’ (Grimshaw. However. The other approach would be to increase the capacity of the transmission channel which would involve increasing the bandwidth. Also. if the frame arrives at the bridge faster than its processing rate. It must translate addresses for protocols using different address structures (addressing discussed later in this chapter). A gateway is the most sophisticated and complex of the facilities that connect networks and allows for different protocols at one or all of the layers of the systems architecture (Chapter 8 is devoted to systems architecture). Unlike a router. structure of format and maximum message size. It must convert the format of the message because it may be different in character codes. the bridge is concerned only with networks that share the same protocols. which shows the cells of an m ð n matrix. which ‘is a device that simultaneously combines the functions of bridges and routers to give the ‘‘best of both’’. In telecommunications we do not change the direction of flow but instead reduce the redundancies in the message. In the case of Figure 4. like the name of a corporation.

This compression problem is of great interest to the movie industry where standards have been developed. a linear programming or input output analysis problem with m D 100 and n D 150 (which is not too large for such problems) and a value of four significant digits. but lest a potential e-mail user is intimidated by the header format it should be stated that an e-mail address format is much simpler. For Message ROUTER Message compressed NETWORK INTERFACE NETWORK NETWORK INTERFACE Decompression ROUTER Message Figure 4. voice. approaches being researched which include algorithms based on fractals and wavelets. it is necessary that it has all the information required to find the destination along with control information. Whatever is transmitted (video. Researchers suspect that a video display with a web of thousands of tiny computers one for each picture element of a video screen might be able to generate video images with far less data than conventional videos currently required. it is divided into packets. Sometimes.5 million for VCR quality. with the message being assembled back into its original form at the other end.4. number of fields and content of the fields depending on the media of transmission and the technique of communications used. The message is compressed before the message is sent and decompressed at the other end. Addressing For a message to be transmitted. If the message is a long one. however. data or text). MPEG1 and MPEG2.Switching and related technologies For a matrix in. the savings would be around 60 000 bytes of data. and each packet has its own header so that the router interprets the header and routes the packets on one of the different paths available. Another solution may lie in combining compression techniques with display technologies. There are. a subject that we will now examine. The headers will vary in format (fixed or variable fields). The problem with the MPEGs is that the encoding is very expensive and time consuming. It could be more complex than in Figure 4. the process of compression is the same as illustrated in Figure 4. say. Compression is also important in the transmission of images and voice where whole blocks of the image and voice being recorded can be empty. The Motion Picture Experts Group (MPEG) has agreed on two standards. an error checking character is attached to ensure that no error creeps into the compression process.4. An example of a header for a frame (a unit of information transmitted as a whole) which has fixed sized fields is shown in Figure 4. This error checking could be part of the header in the address. which have reduced the full video signal from 250 million bits per second to 1.3 Routers and networks 41 .3.

the e-mail address of the author once was KHussain@ix. Note that the fields are not fixed in size but the end of a field is identified by a separator such as @ or a dot. which we now revisit.Telecommunications and networks Control Check (CRC. This traverse between the AC (alternating current) of a digital computer and the DC (direct current) world of telephones 42 is achieved by the RS-232C.500 by CCITT. the infrastructure may not support the fast speeds of new modems like the 28. which is a serial connection consisting of several independent circuits sharing the same cable and connector.4 Field in a packet example. 192 times the speed of the original Bell 103 modem. Modems We discussed modems briefly in Chapter 2 and saw how they enable us to traverse freely between the digital universe of a computer to the analogue world of telephones.6 kbps (kilobits per second). an early international organization for standardization. Note also that lower case is used for the address. e. Cyclic Control Check Message Destination address Source address Frame control information.8 kbps. This is because there is a limit to the size of the name field. the modem with a microprocessor first introduced in 1981 now has speeds of 57. It is shown in Figure 4. The newer PCs come with a built-in modem so that it is transparent (invisible) to the user. We shall discuss the anatomy of an e-mail address later in the chapter on e-mail. the payoff period may be just a few months. Note also that the name is not complete. with an initial missing. Furthermore. one standard for global addressing is X. In fact.g. type of frame Flag Figure 4. some are more than early mainframe computer systems. Also. they are so fast that the other equipment in the system cannot keep up with them including many of the computers that they connect.netcom. The faster speed of the modem may cut the phone bills by nearly half and with the price of modems dropping. The computing power in a modern modem with computers replacing the microprocessor is awesome. . Many public phone circuits (even in advanced countries) have limits in their bandwidth and lose signal quality in long loops (even local loops). One crucial device for transmission and telecommunications is the modem. Sometimes the whole name is also required to be in lower case. There are standards for addressing.5 as a box on top of a computer and can be ‘added on’ to a computer when telecommunication capability is In speed. but suffice it to say that it is fairly end-user friendly. priority level. It is translated into a machine readable equivalent for interpretation by the router. So what else is new with the modem? Why all the interest in an old device that has been around since the 1950s and 1960s? The answer is that the modem has greatly evolved not just in speed but in other capabilities.

with companies overloading their circuits. The V.32 (at 9. The processing capability at both the sending and the receiving ends must be very computer intensive in addition to consuming storage in vast quantities if you want a reasonable resolution of the image. but the downside is that the computer system must be turned ‘on’ and have the fax program in resident memory of the computer at all times. The sending end may also have a mechanism that compresses the data before transmission. This is not a trivial problem as any practitioner in Operations Research and 43 . There is also greater interoperability between the different models of modems from different manufacturers. The V.5 Modem converts signals the switching is often poor. The advances that came in 1984 did not come from industry but from the nudging of the ITU (International Telecommunications Union).22bis and is still in wide use today though it is being steadily replaced by the successor V. Smart and intelligent A smart device is one that has a microprocessor or microcomputer to do its computations. better diagnostics and better documentation. enter the address where it is to be sent. if what you want to send what is not in computer memory. The faxes are strings of 0 and 1 bits that represent a graphical image of the transmitted page which are processed at both the sending and receiving ends by computers.6 kbps) which is currently in wide use. An example would be a router that needs a computer to do all its many computations for determining the optimal or near optimal (or even a better) route for messages. Of course. and off it goes. But the fax in a computer is so easy to operate. Another advance in the modem world is the fax modem which came into international prominence when the government of the USSR had unexpected trouble with its rebels because of the effective communications amongst the rebels through the fax.42 standard is concerned with error control and the V. The new standard was designated the V. You select on the screen what you want to fax. They are more reliable and have good technical support for detecting and solving operational problems.34 standard probes the line to diagnose its qualities and weaknesses. the successor of CCITT. Some computers coming out in early 1996 came with in-built scanners which will greatly increase the use of fax by computer.42bis with data compression.) Modern modems have improved file-transfer performance.Switching and related technologies Includes RS-232C which contains several independent circuits sharing the same cable and connector Compression control Error control Analogue signal Modem Digital signal Figure 4. (The V symbol stands for modem standards while the numbers is part of a block code for modems. then you must have an optical scanning device for input. Fax images by computer (even PCs) are sharper than the regular fax machine.

‘cheaper’). computing capability is embedded in the device which then makes it smart. In our example. It also needs a computer that performs the many necessary calculations. then APC). We are learning much about fuzzy systems and this is being incorporated in many daily household and industrial goods. Deutsche Budespost R O M Converts digital and analogue signals MODEM CHIP R O M CPU Performs error correction & data compression Analogue side Power Supply RS-232C Figure 4. If there were no computer capability. the fax modem is both smart and intelligent. We can also expect optimal models for routing for minimum cost as part of the work being done on optimization in telecommunications (Luss. be it for the airlines. Consider the router and the problem posed in Figure 4. then the modem could be called a dumb modem. What can we say about route 2 compared to route 3? If we applied a decision rule of transitivity (if APB (A preferred to B). This ability to make inferences given data (of preferences in our problem) and decision rules or heuristics (rules of thumb) enables us to make inferences that reflect intelligence and hence the device can be called intelligent. ‘better’. let us assume that comparisons showed that route 2 was better (by a given set of objectives) than route 1 and that route 1 was better than route 3. car rental or the theatre. 1989).6 Layout of a modem 44 . for they have long been concerned with the Travelling Salesman problem. It is going to be much smarter and more intelligent as we learn more about making intelligent choices even if the variables are ‘fuzzy’ and not discrete (like the words ‘good’. But it is the capability of making inferences about sounds and images that makes the device intelligent. since route 2 is better than route 1 and route 1 is better than route 3. As these intelligent and fuzzy systems are adapted to telecommunications we can expect more effective and efficient devices. an intelligent device also needs a computer but makes computations that involve making inferences. In contrast. This exercise in smartness and intelligence is designed to make the reader appreciate the possible future of the world of telecommunications.) Further. (You perhaps thought that it was a trivial diagram but there is ‘a method in this madness’. You may consider this a trivial problem.Telecommunications and networks Management Science will tell you. 1994) which will allow the system to perform logical operations after a call and determine the credit status of a caller. Now suppose that the modem is a fax modem with all the capability of discriminating voices and images. There is even research being done on an AIN system.6 with a computer chip which makes the modem smart. then.1. and BPC. so let us consider a modern modem. one can safely infer that route 2 is better than route 3. ‘high’. which is to determine an optimal route for a salesman that must make a specified set of stops on his sales trip. all in a fraction of a second. To calculate the routing problem. This has important relevance to on-line real-time reservation applications. merely the RS-232C. One is shown in Figure 4. the Advanced Intelligent Network (Brim. This capability requires pattern recognition which uses heuristics developed by practitioners of AI (Artificial Intelligence). BT in the UK.

The airlines have their main maintenance and administrative facilities at the hubs with minimum facilities at the other points of service. There were de facto protocols from IBM and Xerox Corporation and more formal ones from the DOD. The DOD established the TCP/IP. the DOD. protocols that do not favour any one single manufacturer. We shall have more to say about TCP/IP later on in Chapter 8. once you have the necessary devices to transmit and have the transmission media it is time to have protocols. Before we do that. we need to discuss one other important facility: the hub. echo control (largely for diagnostics). But before we get to the wide area communications. Both are concerned with control of transmission but IP/TCP was designed for a specific type of network. Besides TCP/IP and IP/TCP. there are other network protocols such as IPX. instead of serving each point to point directly. TCP/IP is an attempt towards an open protocol and a telecommunications lingua franca. error control. These are rules for sharing telecommunications. Protocols Whether smart. For media other than data. This arrangement may inconvenience the passenger somewhat but it is tolerable if you are not pressed for time and the money savings are worth the delay. It is also the favourite choice of an enterprise-wide backbone protocol where a backbone is the highest hierarchical level. the Ethernet. sharing. the Transmission Control Protocol/Internet Protocol. We need protocols that are ground rules for interaction. DECnet. It is such proliferation of protocols that led manufacturers and designers of protocols to a recognition in the 1980s that there was a need for open protocols. It has options for five protocols: routing control. This would enable one to plug a cable into a wall socket and achieve interoperability of networking. We have hubs in daily life like the hub at some airports. The IP part of the protocol addresses routes and delivers packets to the destination network and host computer. Some airlines have a set of points that they want to serve but.Switching and related technologies Telekom in Germany and American Express in the US are all working on AIN applications. thereby greatly decreasing the overhead costs and increasing efficiency. this is also known as plug-and-play capability. AppleTalk. that is. media and devices on it. a subject that is best discussed along with long distance and wide area networks. and for long distance transmission. the subject of our next chapter. we cannot have two messages going in opposite directions on the same channel! Protocols are conventions for representing data in digital form and procedures for coordinating communication paths. NetBIOS. Basically. routing and distribution in the relaying of information. Hubs An important facility relating to switching is the hub. which are agreed upon rules of behaviour. They had the problem of having a large inventory of computers of different models and different manufacturers that did not interoperate and wanted software protocols that would enable transmission across the many configurations of computing. packet exchange control and sequence packet control. but in telecommunications protocols for information exchange have only emerged since the 1970s. they connect each point to at least one hub and then switch the traffic to the desired point through one or more hubs. In its early version. then you have a special line dedicated to your service. We will discuss backbones in later chapters. If you are pressed for time and have direct traffic. For example. Department of Defense in the US. such as voice or audio. TCP/IP was most appropriate for data transmission and primarily for local connections. the TCP is responsible for assembling packets for transmission and to properly order and reassemble them upon request. and XNS. other approaches are desirable including the ATM. What was needed was the ability by users to mix and match networking hardware and software so as to create customized networks. We have had protocols for telephony from the early days of Morse and Bell. The TCP/IP should not be confused with IP/TCP. Once connected to a backbone there is a guarantee of interconnections. 45 . Soon computer companies in the computer and telecommunications industry started following the TCP/IP or else they could not do business with perhaps the largest user of computing. we need to discuss local communications and local area networks. intelligent or dumb.

all these terms will come together and make sense. while other hubs restrict the type of topology. and gateways 46 . by the end of the journey in this book. Some hubs have a standard backbone. Hopefully. while others use satisfying algorithms and simulation to find a route.7 Repeater. and if so what part. others handle multiple nodes for each port. We have come across the concept many times but dare not use it too often because APPLICATION Gateway PRESENTATION SESSIONS TRANSPORT NETWORK DATA LINK PHYSICAL Router Bridge Repeater Figure 4. There may be no solution to this problem when discussing telecommunication and networks. the author has violated a cardinal rule of not using concepts or terms (layers of network architecture) without having defined the terms. Some hubs handle all topologies. In doing so. e-mail. All traffic must also be analysed and monitored for all segments of the network to identify potential bottlenecks. others do not. Some hubs are fixed-port. There is a similar problem with the term ‘network’. Some hubs allow FDDI links. repeaters. Most hubs are hybrid and have a varying combination of the many features available. Some hubs are modular and others are not. Summary and conclusions In this chapter we have examined the protocols and devices necessary for telecommunications. Some hubs use optimal algorithms such as for routing. All this can be done from one central management point to be monitored by a single console. Some hubs are intelligent in that they have the ability to make intelligent decisions and handle ‘fuzzy’ variables. and so can interconnect conventional network modules on one floor. as is done in Figure 4. An elegant way to summarize the discussion on bridges. some allow ATM links. All traffic must be ideally routed seamlessly and efficiently. congestions and delays. router. but we shall handle the problem by minimizing the terms not yet defined and scrupulously define and explain them later. sometimes even remotely. One may be the store-and-forward but then the hub must decide whether to store all the traffic on a segment to stack up all the traffic or only part of the traffic. Such a hub can have multiple switching and network connections in one box. It has heavy duty LAN switches that can handle heavy traffic with high bandwidth demands and must handle sharing of bandwidth with varying traffic like data. and switch traffic directly at high speed. There are many ways of handling traffic.7. multimedia and real-time like videoconferencing with varying demands of high bandwidth and low latency. bridge. Some facilities are dedicated to certain traffic while other facilities are shared. routers and gateways is to compare the different layers of network architecture.Telecommunications and networks Much the same happens in telecommunications where a hub is where messages are switched to their desired destination. while others do not care which link you use. The hub must also be able to identify and respect priority traffic such as a CAT scan for a dying patient. connect stackable and modular hubs on another floor.

52 65. algorithms.R. The advanced intelligent network: an overview of markets and applications. 25 (2). Stone. Luss. Data Communications. M. Tannenbaum. Data Communications. Govert. In 1995. The new centre has over 19 kilometres of fibre optic cables connecting its hundreds of PCs and workstations with the larger computer systems owned by NASA. S. Halfhill. 241 247.T. (1994). H. has controlled the flights of all the early spacecraft. Prentice-Hall. Byte. Byte. Computer Networks. 47 . 71 86. 24 (12). 21 (2). 57 74. 51 54. AT & T Technical Journal. 33 35. 19 (2). 13 (5). Tolly. and applications. Coping with public frame relay: a delicate balance. Bibliography Brim. J. Telecommunications. LAN interconnections technology. (1995). Even a large and in some ways very important real-time system can be constructed from products that are commonly available and are no longer ‘high tech’.1: Networking at the space centre The space centre at Houston. 19 (9). 23 (2). New techniques for testing switched 56 services. A. Johnson. (1991). 3 6. P. 25 32. These computers and workstations are all interconnected in addition to being connected to the tracking stations all around the world. (1995). Next generation routing: making sense of the marketectures. 43 47. 25 (2). (1994). Grimshaw. The design specification is a commentary on the state-of-the-art of telecommunications and networking. Stalling. 229 241. 113 120. 19 20. (1989). Robinson. (1994). How safe is data compression. PC Magazine. PC Magazine. W. a new command and control centre was partly operational and in its beta phase of testing to replace the old centre and to prepare for the space shuttle into the twenty-first century. 68 (3). When is a fax not a fax. Telecommunications. D. One design specification of this complex and important networking systems was that almost all the equipment must be ‘off the shelf’. Data Communications. V. 13 (15). Saunders.Switching and related technologies it has not yet been well defined or explained. LANs make the switch. Testing TCP/IP software. Optimizing in methodology. Case 4. Bryan. (1991). E. Telecommunications. J. (1992). 26 (1). T. (1989).34: off the starting blocks. (1994). Texas. Telecommunications. K. This was specified in order to keep maintenance easy and not as costly as in the previous centre. We now liberate ourselves from this constraint and discuss the concept in detail as the topic of our next chapter. 28 (12). (1992). (1994). J. Streamlining high speed internetworking protocols.S.

a network is an interconnection of nodes. the characteristics and objectives of networking. as in a building. a company headquarters or a manufacturing plant. It will enable video-conferencing and conveying video images by businesses across national borders. like on a university campus. doing anything and everything that involves computing. the exchange of research results. networking as a paradigm for computing. and the protocols and switching methods necessary for a LAN. protocols and even hardware and switches for directing the transmission along the desired route. Thomas A. or a set of buildings. There are supercomputers manipulating billions of commands per second. or a LAN: an interconnection of computing resources within a limited geographic area. railways and even grids for electrical power. It may even affect our social lives and bring the world closer together if the increase in e-mail (electronic mail) demand is any indication. a Wide Area Network. Stacks Networks connect people to people and people to data. nor does the even faster HIPPI spell the end of FDDI. Computers are handling billions of messages in offices and businesses and helping with design and production in manufacturing plants. In each case. MANs and WANs deserve a separate chapter and will be the subject of our next chapter. forecasting weather. Interconnecting this computing power will dissolve the temporal and geographical barriers and enable us to work and play with partners that are distant and even in other countries and continents.5 LANS: LOCAL AREA NETWORKS Just because FDDI is faster than Ethernet does not spell the end of Ethernet. and a WAN. remote medical diagnostics. The impact on business and society of such interconnections and the problems in global interconnectivity are the subjects for later chapters. or even while walking or jogging. In telecommunications a network is more than interconnectivity of nodes: it is a system with software and protocols that enables the exchange of information and computing resources. and sharing of information and resources that are otherwise isolated and disconnected. 48 Interconnectivity Computers these days are both powerful and cost-effective. They need special software. circuit switching. FDDI. In the early 1980s this was localized. Stewart Introduction In our society we have many networks as in roads. Carl Malamud. a Metropolitan Area Network. This is referred to as a Local Area Network. in the car. suffice it to say that interconnectivty can have profound . working on chemical reactions. Also discussed are the many approaches to access in a LAN including the Ethernet. SONET and the wire-less network. and analysing complex images in medicine and industry. There are over 50 million personal computers with thousands of software packages along with wireless devices based on satellite and cellular systems which enable us to work at home. But there are many situations where the interconnectivity must extend beyond a local environment. They are everywhere. frame relay. The lesson learned from the Token Ring/Ethernet war was that we can have multiple types of data links working together to provide a coherent network. We have a MAN. Chapter 6. In this chapter we will examine the nature and desirability of interconnectivity of computing. token ring.

In addition to sharing the resources of other computers. Many of these devices can be accessed directly. or even another expensive peripheral. there is great certainty that systems will not be merely stand-alone systems. called a ‘client’. This is often true of image processors.1 shows the interrelationship between computers and telecommunications. One such schema for interlinking of processors is shown in Figure 5. and then the end-user. the processors of the same type may also be interconnected to each other in addition to being connected with other types of processors within an organization as well as with processors connected to a LAN. another network can be accessed through a bridge or a gateway. One approach to networking is to have services (including documents or files or a peripheral like a printer) resident on dedicated computers called ‘servers’. accesses the server that has the required service. There is the L A Specialized workstation N Your PC/workstation Other PC/workstation Hand-held computer Figure 5.g. In the real world. Figure 5. voice processors. not only to gain access to a more powerful processor. of course. One such configuration is shown in Figure 5. If not available on one network.LANs: local area networks implications on the way we work and live in the years to come.1 is a bus and of course there are other configurations of networking.1. it is equally desirable that one would want to share peripherals. Another peripheral may well be a storage device like an electronic cabinet or as archival storage device. Networking and telecommunications are one of the fastest growing sectors of the computing industry. capability and even design) into a network that enables the sharing of computing resources by all those connected to the network. Many peripherals are very expensive and a department within an organization or even the organization itself cannot afford to have one of its own and must share. Whatever technology emerges for the future. once they are attached to a network or through another multiplexer or another computer. The configuration for networking used in Figure 5.1 Interconnectivity of a pc/workstation 49 . This configuration is called the ‘client server’ approach and is discussed in Chapter 10 as an alternative organizational option for computing. but also to gain access to another data/knowledge-base or software package. Parallel processor Supercomputer Specialized computer. networking will offer the access necessary for the sharing of processors and other scarce computing resources.1 in order to keep the diagram simple. or even a fast printer. All these possible connections are not shown in Figure 5. Once standards. and preferably international standards are adopted. It is telecommunications that integrates computer systems (often disparate in terms of capacity. In the increasing complex information systems environment.2. e. Physical linking is. One factor that is holding back a greater acceptance and use of networking is a lack of standards. there will be a need for the interconnecting of processors. optical scanners. not sufficient for accessing other computing resources.LISP machine Mainframe Mine To (another) LAN MAN WAN standards that range from architecture to the format of a message.

We will now examine such characteristics. However. 1994: p. So many issues of interoperability in networking are still to be resolved but progress is beginning to be made. 50 except the last one.1. Once this is available. In comparing networking with batch processing one finds that there is little duplication in characteristics or on application appropriateness. and a simplified way to order digital lines. 130). more specifically LANs.’ (Fritz. this is not true .Telecommunications and networks Other ethernets Office workshop Gateway∗ Printer Information processing centre Electronic file cabinet Interface Production workstation Production machine Micro graphic cell Terminal in office Fast printer Ethernet multiplexer Terminals Processor Typing system cable Printer Computer Figure 5. It seems logical then that these two modes of processing will coexist. a network is imbued with certain characteristics that are associated with networks and others that are desirable and depend on the application. There is not yet a good agreement between router-to-bridge unless your router can pretend to be a bridge.2 Peripherals served on a LAN need for software and protocols that facilitate and enable access. It is compared with the traditional paradigms of computing in Table 5. Interoperability requires agreements such as the one for operations between routers. are as follows: ž fast response time ž low delays which should be bounded ž notification of estimated delays when they occur ž high throughput (thruput) ž high channel capacity ž fairness of protocol in assigning access (within a priority scheme) ž ability to add or remove a station easily ž low and fast maintenance ž interoperability The terms used above are common to many information systems and will not be discussed further. Characteristics of networks Characteristics and performance objectives of networks. Networking as a computing paradigm Given the inherited characteristics and some desired performance objectives. ‘With full device interoperability. one can safely say that networking is now a new and different computing paradigm. national and international standards. plug-and-play networking may yet be possible.

Networking therefore emerges as a new and important paradigm of computing. There are many access approaches and they depend on two things: first. not just to other computers and peripherals. the topology of the nodes in the network. It has many applications that would otherwise not be possible. Networking also has many inputs made possible because of its remote connections. including a LAN. but also to other databases. from e-mail to downloading programs Mobile computers Client computers or terminal Other computers Private databases Public databases Satellite info. a network offers much wider access. These are discussed in three later chapters. Here networking has broader scope and extension for sharing computing resources and so time-sharing will be subsumed by networking.LANs: local area networks Table 5.1 Comparison of computing paradigms Batch processing Main target audience Connection User status Objective Operations Applications End-user of output None Subservient Computation Processing in batches acccess Customized reports Time-sharing Access for end-user Telephone Dependent Access Varied processing Shared computing resources Desktop (stand-alone) ‘Owner’ None Independent Computation Varied Varied.3 Inputs to a communications network with time-sharing. and second. 51 . is the access approach for navigating the network. Important to all networking. As for stand-alone computer systems. there is a commonality in equipment access: both use a desktop.g. and other publications Teletype setting Electronic office Home computers Fax machine Figure 5. but often localized Networking Anyone bonafide Telephone and network Unrestricted Communication Remote Varied. weather info Telecontrol Telemetry COMMUNICATIONS NETWORK On-line input devices Newspapers. These inputs are shown in Figure 5. journals. e. Thus desktop computing may well adopt the networking connection. But while the desktop provides stand-alone computing with perhaps a local database. the switching mechanism chosen.3.

4 Basic topologies in networking 52 . It is therefore more reliable. if one side breaks down. configurations formed by connections between devices in a LAN. test. all fails Easily done Ring Bidirecttional path possible By convention Restricted Has an alternate path if one fails Not easily done Routing Control Nodes Robustness Modifications Requires full duplex modem No routing By convention Minimum distance between nodes Failed nodes bypassed Easily done BUS STAR RING Figure 5. especially when compared to the star.4. and has the ability to bypass a faulty node. (There are many combinations of the basic three topologies not shown in order to focus on the basic topologies. the star and the ring.2 Topologies compared Bus Star Centralized Centralized No restriction If centre fails. Circuit switching is what is used in our daily Table 5. the star is easiest to control. The ring also has the bypassing capability but.Telecommunications and networks Topologies and switches There are three basic topologies.) It is generally accepted that the bus is simple to comprehend.2 and in Figure 5. is easy to change (add or subtract nodes). but difficult to change. where if the centre breaks down then the entire system collapses. These are the bus. maintain and manage (since it is so centralized). There are two basic types of network switching methods: circuit switching and packet switching. it has access from the other side. However. unlike the bus. A comparison of the three basic topologies is summarized in Table 5.

Note that a message from origin 1 to destination 6 does not go directly through A and E but with a diversion through A. you have the exclusive use of the circuit and can stay as long as you want.6. with all the 5 6 E D 1 A C 4 2 B From 1 to 6 via A. Circuit switching is appropriate for voice messages and where human interaction is required. but it is not appropriate for the transmission of data as needed for most computer processing. and getting off precisely when it is so necessary requires considerable human cognition and attention.5 Circuit switching 53 . Bursty traffic has a high variation and unpredictability in transmission rates (it varies from 100 bps for a terminal to a million bps for many a processing job). packet switching is most appropriate. you hang-up and break the circuit connection. Because of the importance of the connection. Circuit switching is shown in Figure 5. C and D Figure 5. The connection he got was through New York. and that is the end of it. Circuit switching is illustrated in Figure 5.5 using a very simple segment. you hangup. Each packet also has a destination address which is read and used to route by a packet switch which could be a programmable computer. B. D and E. Here. Thus. After you try for a connection (if you get it). with bursts or surges of data sent at high speeds for short periods of time. Here.LANs: local area networks telephones: you dial. This indirect path may be the only available path and is what often happens in our daily telephone conversations. especially if the traffic is heavy and the possible exits are many.5. When you are finished. E From 2 to 5 via A. the connection need not be direct but whatever is feasible at the time. being in the correct lane for the desired exit. a segment of another net for packet switching is shown in Figure 5. That does not matter much because each packet is uniquely identified and can be reassembled at the other end. you converse. the traffic is bursty. a message is broken up into packets which are units of information travelling as a whole between devices conceptually similar to a bus. For such traffic. Cars for different destinations share the same road until such time as their exit comes and then they leave on another road towards their destination. The packets share the transmission channel and may not all go together. Your author once had to make a phone call from Paris to Marseilles in the south of France. D. Knowing where to get off. Fortunately. in telecommunications the switching is all automatic. The packet switching process is analogous to a motorway with cars representing packets. this approach is called connection switching.

depending on what links are available at the time. The protocols needed for packet switching are contained in X.3. This does not happen in circuit switching where a message can be lost when facing heavy traffic and congestion. Packets switches (programmable computers) Packets for host 1 Packets for host 2 Packets for host 3 Packets for host 4 1 Figure 5.Telecommunications and networks PACKET SWITCHING 3 2 4 Legend Clients to host (shown only for host 4 for sake of simplictiy).25 network.25 which is approved by 54 CCITT. The X.3 Summary of circuit switching and packet switching Circuit switching Packet switching Is connectionless Each message is one or more packets Logic for control at each node Channel is shared and more than one message is moved simultaneously Not interactive Data is bursty Can be‘store-and-carry’ Requires a connection Each circuit call is one or more messages Logic required at switching centres Circuit is for exclusive use Interactive Data can be a continuous stream No ‘store-and-carry’ decision-making and choices being made by the network software.7. . packet switching may take different paths from origin to destination. If the traffic is very busy. Packet switching and circuit switching is summarized in Table 5. the message is delayed and the system will store and forward the message when passage is possible.6 Packet switching Table 5.25 protocol was so important and dominant that packet switching was often referred to as the X. Error control is summarized in Figure 5. Being connectionless.

LANs: local area networks


Checking for errors done between each pair of nodes before proceeding further


ORIGIN 1 2 3 4


Checking done only at the end-station (destination, 5)

Figure 5.7 Error control in frame relay and packet switching

Access methods
There are many approaches to accessing a destination in a network. It is a combination of topology and a switching method. Each is selected for the appropriateness of a topology and a switching method for each set of applications. We shall discuss these approaches starting with the oldest: the Ethernet.

more than a mile. If another transmission is detected then transmission ceases and tries again after a random interval of time. This approach is called the SCMA, Carriers-Sense-Access Method.

The token ring
Another approach is the token ring which combines the ring topology with packet switching method. Before transmitting a message, a token (short series of bits) is passed. If accepted, then the computer accepting the token is free to transmit one or more packets and other computers wait until they get the token. This avoids clashes and establishes a mechanism for sharing a transmission media of cable. There are many variations of the token ring approach including the Cambridge Ring developed by the University of Cambridge in the UK.

Ethernet is a combination of packet switching on a bus topology and was first used by ARPANET, the network implemented by the Department of Defense in the US to facilitate communications between its many researchers in government business and academia. In parallel with ARPANET, similar projects were started in Europe, especially by the National Physical Laboratory in the UK and the Institut d’Informatique et d’Automatique in France. In 1970, the Research Center for Xerox announced the Ethernet standard which has since been adopted by many private and public networks all around the world. Thus, it is not only the oldest LAN system, but perhaps the most used network system. In an Ethernet system, the network broadcasts a signal over a coaxial cable over a distance of

Circuit switching
Circuit switching approach creates an end-to-end path before invoking the flow of data between two nodes. It is a simple technique and used extensively by the PT&Ts and other telephone carriers. However, it is not always the most efficient approach since setting up a connection may take 55

Telecommunications and networks longer than the message itself to which many of us making a phone call can attest. The frame relay can be used as a high capacity backbone for the X.25 access networks and for LAN interconnectivity. Its advantages are: ž ž ž ž high speed interconnectivity are lower costs; minimum delays since the network does not terminate protocols; no significant new software by bridges, routers and gateways; and evolutionary path for high-speed LAN interconnection services (Bushman, 1994: pp. 42 3).

FDDI, Fibre Distributed Data Optical Interface, combines the token ring approach to the high capacity of fibre optics achieving high rates of up to 400 million bps. The system is organized with two parallel rings, so that if one fails, recovery is made on the other ring. This approach is also called the Dual Ring Approach.

Frame relays differ from packet switching and is best described in the comparison of Figure 5.7.

Frame relay
A frame relay uses packets in a circuit switching environment. Furthermore, it is similar to a virtual circuit which delivers packets in order but with variable delays, except that virtual circuits are determined at the time subscribers are connected to the system. Frames are 64 1500 bytes in length. The start and end of a message is identified by a flag. Frames can also handle video and voice traffic. They are popular partly because they are attractively priced. Both the frame relay and the X.25 for packet switching, when of variable length, had to constantly adjust the flow and timing of messages. However, with reliable digital circuits in the late 1980s, designers stripped off many of the X.25 functions, reduced the overhead, and subsumed the X.25 into the frame relay. Firms that are new (relatively speaking) to telecommunications, like MCI and Data Communications, do not even offer the X.25; instead they rely on the frame relay. Frame relay provides faster service than X.25 and at the same time provides much of the same communications facilities including flag recognition, address translation, recognition of individual frames, and filling in the interframe times. Some of the benefits of the frame relay are: ž ž ž ž 56 higher network productivity; reduced network delay; savings in bandwidth; better and more economical hardware implementation.

SONET is the acronym for Synchronous Optical Network. It supports a multiplexed hierarchy of transmission on speeds ranging from 51 to 2400 bps. The SONET system ‘allows data streams of varying transmission speeds to be combined and extracted without first having to breakdown each stream into its individual components.’ (Cerf, 1991: p. 78). The SONET is used by virtually all important carriers except perhaps the largest, AT&T, partly because SONET offers its own optical algorithm. It offers superior performance and monitoring and can route around network failure points in 40 to 60 milliseconds. SONET may well be the gigabit backbone for networks, especially long distance wide area networks, the subject of our next chapter.

Wire-less networks
Wire-less LANs are capable of transmitting 10 Mbps within a room or building. It is similar to cellular technology for data transmission. Cellular phones rely heavily on analogue broadcast techniques, so using them to move digital data will be inherently difficult. One approach is the CDPD, Cellular Digital Packet Data. The CDPD is similar and yet different from the circuit switched network. A comparison of the two is shown in Table 5.4.

LANs: local area networks



Figure 5.8 LAN as a subset of other networks

Table 5.4 radio

Circuit switched vs. cellular digital packed Circuit switchingCDPD No Yes Each session Dedicated Large transactions File transfer e-mail Yes No Once a day Shared Small transactions Dispatching e-mail messaging

Characteristics Store & forward Large data files and fax Host log on Cellular channel Application profile

among the end-users, we have seen a great demand for LANs. However, this demand will soon require that the boundary limitations of the LAN be relaxed and extended to a MAN and a WAN to provide remote services. The LAN then becomes a subset of a MAN/NII (National Information Infrastructure)/WAN. This relationship is shown in Figure 5.8. We shall explore these possibilities in later chapters. First, however, we need to discuss the MAN and the WAN, the subject of our next chapter.

Case 5.1: A network in the UK
Racal Electronics agreed to buy British Rail Telecommunications for £132.8 million, ‘giving the company a data link passing through most major British cities along British Rail tracks’. British Rail accounts for about 80% of the optical fibre network’s customers. But Racal said, ‘usage at present was only 20% of the network’s capacity’. The acquisition gives Racal strength in the field of voice and data communications, its core activity that accounts for more than half of the company’s revenue. Racal now is, in the words of its Chief Executive, ‘a carrier’s carrier.’ Source: International Herald Tribune, Dec. 6, 1995, p. 17.

Summary and conclusions
In this chapter we examined the need for connectivity and examined how connectivity leads to networking as a new and important paradigm in computing. We also noted that there are three basic topologies: the bus, the star and the ring. Also, there are two basic types of switching: the circuit switching and packet switching. Each has its own advantages and limitations. An application environment would make it desirable to combine one topology with one switching method. This gives us a variety of methods of navigating the LAN: the Ethernet, the token ring and its variation the Cambridge ring, FDDI, frame relay, wire-less, and the SONET. Each is briefly described and a comparison is made. An important approach is the ISDN which is discussed in a later chapter, Chapter 7. With the proliferation of computers and the increase of computer literacy and experience

Case 5.2: The Stentor network in Canada
Stentor is an alliance of Canada’s major telephone companies with integrated local and long 57

Telecommunications and networks distance services. It also offers seamless services with North America through its alliance with MCI. Stentor maintains the world’s longest, high density and fully digital fibre optic network stretching approximately 4800 miles across Canada. It services include digital switching, intelligent network services, high reliability (second to Japan) and a 58% telephone penetration. ‘In 1994, Stentor announced the Beacon Initiative, a project that will bring the information highway to 80 to 90 percent of Canadians by 2005.’ The activities of the Bacon Initiative will include a $8 billion upgrade of the telephone networks over the next 10 years; a 500 million enhancement programme over six years to provide national interconnectivity; the creation of a new multimedia company that will provide and distribute multimedia services; and venture capital for the development of multimedia applications and products for the information highway. Source: Stentor, Dec. 1995, p. 3. The servers are to be provided by Control Data Systems Inc. whilst the network service provider is Rayes Technology Co. Source: Computerworld, Nov. 6, 1995, p. 76.

Abeysundara, B.W. and Kamal, A.E. (1991). Highspeed local area networks and their performance: a survey. ACM Computing Surveys, 23(2), 221 261. Boyl, P. (1996). Wireless LANs: free to roam. PC Magazine, 15(4), 175 202. Bushman, B. (1994). A user’s guide to frame relay’. Telecommunications, 28(7), 42 46. Brueggen, D.C. and Yen, D. (Chi-Chung) (1990). Local area network connectivity. Computer Standards & Interfaces, 11, 103 114. Cerf, V.G. (1991). Networks. Scientific American, 265 (3), 72 81. Derfer, F.J. Jr. (1992). LAN Fundamentals, Part 2. PC Magazine, 11(7), 229 250. Francis, B. (1991). Linking LANs with laptops. Datamation, 37(10), 61 63. Gareiss, R. (1993). Tommorrow’s networks today. Data Communications, 24(13), 55 65. Gifford, J. (1995). Wireless local loop applications in the global environment. Telecommunications, 29(9), 35 37. Lane, J.L. and Upp, D. (1991). SONET: the next premises interface. Telecommunications, 25(2), 49 52. Miller, A. (1994). From here to ATM. IEEE Spectrum, 31(6), 20 24.

Case 5.3: A network planned for China
The People’s Republic of China is planning a network infrastructure that will have mail-hub servers in twelve of its largest cities by the end of 1996. The country-wide network will connect on-line many corporate and governmental groups with one another and the rest of the world with e-mail.


No technological imperative determines how to link LANs or even whether to do so. Ben Smith and Jon Udell, 1993 If you want to connect Ethernets to ATM, you may need a Ph.D. to figure out the long term implications. Paul Strauss, 1994

LANs in the 1980s were very successful, but its inherent constraint of being confined to a local area is its greatest limitation for the 1990s. While in the 1980s, industry matched resources to applications, in the future, applications may well have to be matched to networks. Future applications are moving away from the legacy applications of processing data to the processing of real time data, text, voice and images. Processing needs are shifting from the desktop networking to enterprisewide strategic computing; from store-and-forward computing to strategic computing; from off-line computing to real-time computing; and from local processing to remote processing. Networking is being transformed from the early environment of military systems tied to private networks that were switched mechanically to analogue and digital devices of today with broadband fibre optically transported messages that are electronically switched by sophisticated software and intelligent network systems. Businesses are becoming less centralized and more distributed. Businesses are no longer operating exclusively within their national boundaries but going global where physical boundaries are no longer an issue. Traffic in telecommunications is shifting from a LAN to a metropolitan MAN and to a wide area WAN. We must avoid a communications gridlock if we are to benefit from open and worldwide markets. To fully benefit from the information age we must have networks that interconnect the millions

of computers and thousands of computer installations world-wide to enable useful and meaningful information exchange. A step towards this goal is a step beyond the LAN and towards the MAN and WAN, the subject of this chapter. In this chapter we will examine the nature of processing in a MAN/WAN and compare it with the LAN. We will also examine the planning and performance management of a WAN which includes bandwidth management and switching management. The discussion of the management of switches will lead to a detailed discussion of the ATM.

MAN stands for Metropolitan Area Network and is a computer network that typically covers a part or all of a metropolitan city. It usually encompasses a compact area with an area that ranges from one to a few dozen miles in radius. In contrast, a WAN (Wide Area Network) is for a much larger area, much larger than either a LAN or a MAN. Typically, the distance may be 100 or a few thousand miles in radius. In practice, a WAN may cover a nation, a continent, and be international and world-wide. A MAN is mostly copper and fibre cables like the digital circuits based on the local phone company and connected to LANs with cable or with microwave connections to a nearly microwave system. The MAN is not heard of much these days perhaps because it mostly uses a stable and proven technology developed by the telephone 59

We can see the demand driven technology in the area of multimedia where the end-user is no longer satisfied with printed reports in batch as in the 1960s and 1970s but demands graphic and image processing especially in the factory with CADD (Computer Aided Drafting and Design). in addition. . With each inclusion of media there is a demand for more and better service and for these applications to be integrated. This may not be a common scenario today. plug the computer into the wall socket. Output in printed form is no longer sufficient.1 Interaction between demand and technology demand for voice processing. and are connected to the world through networking. voice. but in the sense that it is (potentially at least) equipped by computers. That is why an integrated digital system is of such interest and is the subject of a separate chapter. With each reduction of response time there is demand for more reductions and faster processing. ž the extent of remoteness of processing is continuously enlarging to include the entire world. ž an increasing number of end-users and organizations using networks. Each room is wired for plug-andplay equipment. ž increasing need of multimedia processing including data. worker and the home-owner. The end-user cannot tolerate a waiting time of even a fraction of a second.1. To add to all this. All this must be done quickly and seamlessly (smoothly) across remote distances. output should be as voice with a possible input response also as voice and all in real time. You acquire your hardware and software. This cycle is shown in Figure 6. The demand drives the computer industry which in return drives the telecommunications industry. one must consider some of the facts of the changing environment: ž an ever increasing number of computers and workstations with regular upgrading of technology and all of them quite powerful. And then that too was not sufficient and there is now a demand for audio processing. One can safely say that in our world the need for computing and telecommunications is demand driven. It is of special interest to the computing professional because it is an analogue signal whilst computers are digital. images and video with possibilities of films-on-demand in the near future. So it is important. the level of aspiration is steadily nudging up with the traffic mix being very multimedia and in real time as shown in Figure 6. With each advance there is demand for more advances. it is part of video processing (and voice processing) and sometimes part of real-time processing. The cycle is complete in that the telecommunication industry drives the computing industry which in turn influences the demand.2 but it is more important in relative terms. But soon that was inadequate and there was a 60 Demand on telecom industry interconnectivity broadband capacity multimedia transmission Figure 6. Thus. Chapter 7. So. an important basis of the so-called intelligent building.2. both analysis and synthesis. It is important as a stand-alone technology and has many applications for itself. ž the end-user certainly able and desirous of end-to-end interconnections and integration of applications. Voice is shown as only one of four media in Figure 6.Telecommunications and networks companies. there is a demand that this be done in real time with animation and in colour. which includes the manager. intelligent not in the AI (Artificial Intelligence) sense. ž increasing complexity of businesses that demand remote processing for their input or output or both. Demand for applications by end-user voice & video processing teleconferencing cinema-on-demand Demand on computing industry multimedia workstations Planning for a WAN In planning any network. the demand on the computing industry and on the telecommunications industry to transport all this information is steadily increasing. however. It is. ž an increasing awareness and increase in computer literacy of the end-user. but many professionals in computing see that is a viable scenario.

These strategies are all of what is called bandwidth management. which we shall discuss below. reducing the load on bandwidth without reducing traffic (by compression Normalized delay 10 perhaps).3 Link utilization and delays (adapted from Weiss. This includes the strategy of switching such as the hub and spoke method and of course the reliability and robustness of switches that can operate reliably in heavy load as well as in a variety of traffic conditions. One important performance measure is to reduce delays. One approach may be to have a secure system with low losses and high reliability.3 5 2 Utilization 50% 80% 100% Figure 6. Bandwidth management There is considerable empirical evidence which supports conventional wisdom that delays increase as the system gets loaded and its utilization increases. and whether we talk about a MAN or a WAN. 1990: p. by adding additional bandwidth. This is shown in Figure 6. This is a subject that we will discuss in Chapter 12. Not to scale 61 . or by increasing the speed of the links.MAN/WAN Text Images Voice Data Real time Figure 6. This can be done by increasing bandwidth. on the Security of Networks.2 Traffic mix in telecommunications Performance of a MAN/WAN Whether we discuss voice processing or not. 58). We shall also discuss another reason for delay and even loss of reliability. But what is performance for telecommunications and networks? It depends on the perspective. and that is switching management. we must design and implement a system that has good if not high performance. but both end-users and technicians will agree that delays are important.

as in the US.5 Mbps which is the speed on some WANs. The problem is relatively easy when one considers just one type of traffic: data traffic. The ATM ATM stands for Asynchronous Transfer Mode where asynchronous means that the signal is not derived from the same clock. funnel traffic on small planes to large hubs and then transfer them on non-stop long hops across the ocean sometimes taking up to 12 hours. Yet another approach would be to increase the speed of the transmission link. This is analogous to having not just auto and car traffic but also having train and waterway traffic all at the same place. images and video with some in real time. trains. Part of the solution lies in a robust switching technique. Some airlines. This complexity increases not just with the increase in traffic volume but also with the change in traffic mix. Such filtering also eliminates unwanted network traffic. Complexity increases very quickly. and therefore does not have the same fixed timing relationship. But now consider a mix of traffic: data. the technical problem is that such switching at hubs gets quite complex with disastrous results when timings are not just right. We have a similar problem in air traffic. assuming 100% efficiency and no contention Link speed (kbps) 9. to be discussed in a later chapter). It was partly in response to the need for large capacity and partly in response to the need to handle voice and video in telcommunications that the ATM was developed. a canal and pedestrians come together at the same time and yet all the traffic flows seamlessly. The moral is clear: keep the local traffic on slow and cheaper channels but transfer the long-haul loads on the faster channels even though they may be more expensive because they are more cost-effective. Multilink is software-based while bonding is hardware-based. The designers of Multilink did not anticipate this problem arising and have no good solution for it.Telecommunications and networks Table 6.1 where the transmission time can drop from 5460 seconds to 3. WAN. or to have all the local traffic funnelled into a hub and then transmit the load over long distances on a high speed 62 . In packet switching we can see the parallel of packets (passenger) sharing the same channel (lane of road) and then being shunted off when they must make a connection.6 kbps (which is the speed of many modems on a LAN) to 1. The problem in telecommunications must also be solved not just for a mix of traffic but for heavy traffic that can well be expected in the days to come. A quick answer to this problem would be to add another channel. This actually happens at Slossen in Stockholm where roads. voice.5 Source: Adapted from Weiss (1990: p. In earlier chapters we have drawn the analogy in telecommunications to highway (motorway or autobahn) traffic with ramps and instructions for making changes at intersections. For an enterprise network operating with a wide variety of protocols filtering can add up to 25 30% more effective bandwidth. Another set of approaches is to reduce the load on the system by compression and the elimination and control of unnecessary data transmitted. This can produce a dramatic effect as shown in Table 6. 58) where we can see that beyond a 50% utilization delays increase rapidly and increase much more rapidly after a 80% utilization. TCP/IP accepts packets out of order and rearranges them in proper order. It offers a high data rate (Gbps) as well as a low latency rate (latency is the time between access time and transfer time).5 seconds by an increase in the link speed from 9.1 Link speed and time required for a 640 kbyte load application. Besides the unhappiness of some passengers. Multilink or bonding (sometimes also called bandwidth-on-demand) are ways to add channels to a network.6 56 112 1500 Transmission time (s) 5460 960 47 3. Multilink does pose a problem: it may deliver packets out of sequence. ATM is supported internationally and in 1988 it was chosen as the switching and multiplexing technique for B-ISDN (Broadband ISDN. and this is where we see the arrival of the ATM. And why is handling of voice and video different from data? Switching management One problem is large switching problems like those faced by a WAN and whether or not to have many point-to-point connections with few switching in between.

The ATM technology provides a common format for bursts of high speed data and the ebb and flow of the typical voice phone. only changes in the image are sent. The header also contains control bytes as shown in Figure 6. periods of silence can be edited out without any loss in the message.MAN/WAN For one thing. the processing is different. ATM uses the cell structure which has a fixed message field of 48 bytes and a header of 5 bytes. Header = 5 bytes ATM is a connection-oriented technology so that each cell is specified before the connection is made.4.5 which Payload (message-data/information) = 48 bytes M E S S A G E Header Check Sum (8 bits) used for control purposes Cell Loss Priority (2 bits) used to indicate whether a cell may be discarded during periods of network congestion Payload Type Indicator (2 bits) used to distinguish between user cells and control cells VCI / VPI field (24 bits) used for channel identification and simplification of the multiplexing process Generic Flow Control field (4 bits) used to control the amount of traffic entering the network Figure 6.4 Structure of an ATM cell ATM structure: HEADER 5 bytes P A Y L O A D (MESSAGE) 48 bytes (fixed length) Frame structure: Payload (message) is of varibale length Header (2−4 bytes) Flag Figure 6. And with video.5 Comparison of ATM and frame structure 63 . With voice. Also. voice and video must be done in real time in order to avoid any losses in synchronization. the rest is already available and can be used for reconstruction of the message. The header in the cell contains all the information a network needs to relay the cell from one node to the next over an established route. This ATM structure is compared with the frame relay structure in Figure 6.

Up to 65 536 virtual channels can be multiplexed into a virtual path.6. voice and video. (Lane. 31). The dedicated connection with the ATM switch which routes messages and controls access in the event of contention. In industry. Connections can also be virtual as in PVC (Permanent Virtual Connection) where parameters (but not necessarily routes) are established in advance in contrast to the SVC (Switched Virtual Circuit) which provides resources as required. More such applications can be found in scientific research with high definition three-dimensional images in real time. they also pay only for the cells they send. ATM ‘users send bursts of as many or as few cells necessary to transfer their data. in contrast.6 An ATM switch 64 . This traffic on an ATM has often been identified as being heavy because of the nature of the traffic. In contrast. computer aided design and ATM To ATM network switch Computer Server Figure 6. These connections are termed ‘virtual’ to distinguish them from the dedicated circuits PCs or workstations used in the STM. 1994: p. 43). Synchronous Transfer Mode. not for the speed of a dedicated facility they may use only part of the time’. 1994: p. can be implemented completely into hardware which is faster. ‘It is the virtual nature of ATM services that will provide greater efficiencies in the future. We have characterized ATM as a connectionoriented technology. Fixed length packets. ATM establishes virtual connections between each pair of ATM switches needed to connect a source with a destination. a recent book written by your author of the rough size as this book occupying over 12 million bits. This arrangement is analogous to the PBX (Private Branch Exchange) used for voice calls. most of its time is spent listening. in addition to connection to a computer that may be a server. Images are data intensive. The header is larger because of the complexity of the variable length message that follows the header but also because it provides data that the software needs to recognize where the message starts and where it ends. 43). Today. (The flag in a frame is a series of bits that indicates the start and the end of a payload message. we shall examine some applications for their data intensity. To give the reader an appreciation of the magnitudes involved. The bursts are sent through a centralized ATM switch that has dedicated connections to its end-users who may have a PC or a workstation.’ (Lane. We start with text. examples of such images would be in CADD.) In Ethernet. conflict when more than one user wants to use the same resource simultaneously. the packet for switching is 64 bytes long. This is illustrated in Figure 6. most communications capacity is idle. which includes images. a full diagram like many in this book. A voice circuit is only 30 40 percent efficient.Telecommunications and networks has a smaller header and a variable length message field. 1995: p. but that does not preclude it from handling connection-less traffic as well. would take around 1 million bits each. An example from the real world would be a medical imaging application such as the rendering and transmission of a diagnostic X-ray would involve 2 10 billion bits of information (Vetter.

LAN Distance: For a few km WAN For over 1000 km Speed: Low speeds High speeds (in Mbps) (in Gbps) Error rate: Low Can be high Ownership: At firm’s level Higher than firm’s level Administration Costs: Less than WAN More than LAN Maintenance: Less complex More complex than LAN Routing algorithm: Simple Complex Switching: Frame Frame FDDI FDDI ATM 65 .MAN/WAN drafting. multimedia networks . ATM cannot deliver ubiquitously across the network in a failure-proof mode as yet . Interoperability with existing technologies such as frame relay and the X. There are many commentators and practitioners who are enthusiastic about ATM and there are some that are more cautious. In summarizing the discussion of ATM. . its promise of high speed. . . images and text all in a time-sensitive environment. the LAN for short haul and the WAN for long haul telecommunications. 1995: p. . . In business. Also. We tend to confuse ATM in teleporcessing with ATM in banks. the topic of our next chapter. one can compare it with circuit switching as in Table 6.2 Comparison of circuit switching and ATM. Another approach is the ISDN. 29). image archiving. Also. the traffic load on networks is heavy not just because of the type of application but also because of high volume. Circuit switching Traffic type: Structure: Delays: Bandwidth: Switching: Orientation: Data Variable length Variable Can be long Wasted Done in software Connection ATM Data and voice Fixed length Fixed Very low Efficient Done in hardware Connection and connection-less ATM will not be possible for some time yet . We shall quote both sides below: ATM is often described as the technology that will allow total flexibility and efficiency to be achieved in tomorrow’s high speed. highly complex connection-oriented networks . Automated Teller Machines. and group work that mixes voice. The problem that many of us. (Vetter and Du. Then there are other video applications in the entertainment industry and in education. 69 70). large amounts of data on simulation have to be moved quickly from graphical workstations to the field offices. Table 6. .3. 1995: pp. They are compared in Table 6. All must now be handled by LANs and WANs.25 is imperative to a smooth and cost-effective evolution of ATM. .3 Comparison of a LAN and a WAN. . Another way of looking at a LAN and a WAN is that the LAN is an access layer which interfaces other networks at low speed and the WAN is the backbone layer of networks providing high performance connectivity. Many of the current network management platforms lack the performance features and sophistication required to manage ATM’s high speed. .2. such as in oil prospecting. . Even ordinary applications like file transfer and e-mail have risen exponentially. mulitiservice. have with ATM is not with the structure or its functions but with its name. especially laypersons. In the office there is teleconferencing. integrated services and universal connectivity . Summary and conclusions A MAN and a WAN are extensions of a LAN. one can safely say that ATM is a proven technology but that its applications lie more in the future (like teleconferencing.7 where the two layers are represented by ‘clouds’ as boundaries. time critical applications to reach the desktop. films-on-demand) than in current reality. an enterprise network entirely composed of Table 6. (Federline. There is more interest in the two extremes. . ATM is one approach to the demands of transmitting data and voice. the technology that finally enables high-bandwidth. . most enterprise users will implement hybrid ATM/frame relay and connect to a device that is terminated in the enterprise network on an ATM port . This relationship is shown in Figure 6.

is organizationally related to Livermore Labs. unless users fully appreciate and are prepared for the consequences of some of the interconnectivity systems that they implement. .7 Access and backbone layers Smith cautions us about the issues to be addressed by the WAN backbone: address capacity planning issues as well as fault reporting and route planning functions. ATM will provide for high performance networks. (Smith. Case 6. As computing power increases. New Mexico. The limitations of existing bus and ring LANs.Telecommunications and networks ACCESS LAYER BACKBONE LAYER Figure 6. Internet working devices. Another factor supporting the use of ATM in a LAN is that LANs are increasingly managed in a centralized fashion by hubs. the demand for higher bandwidths. and Budwey and Salameh look at the role of ATM as occupying a niche in networking. 35). frame relay for lower end applications. overloaded. and SMDS will be the LECs switching interface of choice. (Budwey and Salameh.1: ATM at sandia Sandia Labs in Alburquerque. If priced correctly. In particular. 26). 1994: p. In the global networks. 40). in 66 . interfaces to public networks. Tomorrow’s LANs must operate in an environment in which computing devices are so inexpensive and readily available that there are hundreds or even thousands in a typical office. while ATM will support Gbps speeds. Router management systems focus very heavily on the management of routes/paths and are typically weak in the wide area performance reporting and management . (Wang. 1995: p. LAN internet working needs an efficient WAN behind it for users to achieve productivity improvements. 1992: p. but also ATM switches as predicted by Vetter and Wang: The bandwidth of traditional LANs is usually on the order of tens of megabits per second. With such large numbers of devices any attempt to interconnect them with traditional shared-media LANs would be impossible. (Vetter. and poorly managed. especially for video and other highend applications. and a switched LAN interconnection is seen as an extension (or a replacement) to existing hubs. SMDS may provide the infrastructure of the future global network. their WAN resources could become expensive. SMDS and frame relay will prevail in the next five years as the driving technologies. direct connections of high-end workstations to a centralized switch is seen as an attractive. larger user populations are major reasons for the growing interest in ATM LANs. Increasing use of twisted-pair and optical fibre media foster centralized hub connections as well. Today’s LAN also lacks scalability. . To meet the high load demands and variety of traffic we need an ATM. the expected growth of video-related applications makes the connection-oriented ATM technique a suitable choice for this usage. The LAN of the future will not only have hosts. But this does not mean that it cannot be used in a LAN. 1995: p. 55).

MAN/WAN California, some 1800 km apart. They were one of the first to have a supercomputer and be on the ARPANET. They both are involved in research and development for the US government defence and energy departments. They serve thousands of scientists and engineers and have some of the most powerful computing power in the world. Sandia has 200 LANs and there are 25 in Livermore. These facilities are to be consolidated for purposes of implementation, security mechanisms, reliability figures, meeting performance requirements, and adherence to networking and communication standards. Supercomputing at Sandia and Livermore has two computing environments: ‘secure’ for classified work and ‘restricted’ for unclassified work. Both environments provide TCP/IP (Transmission Control Protocol/Internet Protocol) access for remote FDDI (Fibre Distributed Data Interface) (also used in the UK Parliament in London), or Ethernet LANs to centralized network resources such as graphic servers and mass data storage. Access to supercomputers is provided at both sites by Ethernet/FDDI and other routers. ATM switches are used in the consolidation link based on Cisco routers that provide the connection from the LAN technology to the SMDS/ATM (Switched Megabit Data Service/Asynchronous Transfer Mode) to MAN (Metropolitan Area Network) and using the ATM switches to connect the MAN to the WAN (Wide Area Networks). The measured round trip delay time of the end equipment is 7.1 seconds. Future plans include ‘the possibility of migrating from SODS to the emerging standards for transporting IP and other protocols directly over ATM switches’. Source: Neagle et al. (1994). Developing an ATM network at Sandia National Laboratories. Datamation, 28 (2), 21 3. reduced their administrative staff by 15%; and improved their customer care index by 10%. In addition to increasing productivity and reducing costs, there is also often an improvement in security and quality levels as well as a reduction in delays. The main factors that influence the LAN/MAN/WAN connectivity solutions are the size of the MAN/WAN, the location of the access points, number of users, patterns of traffic, applications portfolio, inter-LAN software, and hardware compatibility. The LAN/MAN/WAN connectivity can lead to an enterprise solution that, according to Chris Gahan of the British carrier BT, ‘. . . is a blend of services, using individual technologies to their best advantage and having the flexibility to change the blend as business changes’. Source: International Herald Tribune, Oct. 4, 1995, p. 14.

Supplement 6.1: Wan technologies
Frame relay Transmission mode: Usage: Variablelength packets Data, some voice 56 kbps to 1.5 Mbps ATM Fixed length, 53 byte cells Data, voice and video 1.5 to 622 Mbps ISDN 48-bit packets SMDS Fixedlength, 53 byte cells Data


Data, voice and video 144 kbps 56 kbps to 34 Mbps

Source: Computerworld, Oct. 16, 1995, p. 69.

Supplement 6.2: Survey on WANs
Focus Data, an independent market research firm conducted a survey of users of Network World on usage and selection criteria of WANs. The results are shown below: Analogue dial-up ISDN Switched digital Other Don’t know 53.1% 33.7% 15.3% 32.7% 9.5% 67

Case 6.2: Navigating LANs/WANs in the UK
LANs, MANs and WANs are being used in the UK for resource sharing and as a mechanism for improving the personal productivity of office workers, as well as for promoting collaboration and work group computing. Britain’s PA Consulting group have found that businesses have increased their sales by 25%;

Telecommunications and networks Based on a possible score of 5.0, the scores for the top five selection criteria used are as follows: Ease of use for remote users Throughput performance Support for a specific LAN protocol Ease of support Management tools 4.56 4.45 4.37 4.12 3.76
Bryan, J. (1993) LANs make the switch. Byte, 18(6), 113 132. Budwey, J.N. and Salameh, A. (1992). From LANs to GANs. Telcommunications, 26(7), 23 26. Fritz, J. (1994). Digital random access. Byte, 19(9), 128 132. Hurwicz, M. (1997). In search of the ideal WAN. LAN, 12(1), 99 102. Kim, B.G. and Wang, P. (1995). ATM networks: goals and challenges. Communications of the ACM, 38(2), 39 44. Lane, J. (1994). ATM knits voice, data on any net. IEEE Spectrum, 31(2), 42 45. Miller, A. (1994). From here to ATM. IEEE Spectrum, 31(6), 203 204. Pugh, W. and Boyer, G. (1995). Broadband access: comparing alternatives. IEEE Communications Magazine, 33(7), 34 46. Richardson, R. (1997). VPNs: Just between us. LAN, 12(2), 9 103. Smith, B. and Udell, J. (1993). Linking LANs. Byte, 18(12), 66 84. Smith, P. (1994). Reconciling the LAN vs. WAN bandwidth management mindset. Telecommunications, 28(3), 51 55. Vetter, R.J. (1995). ATM networks: goals, architectures and protocols. Communications of the ACM, 38(2), 39 44. Vetter, R.J. and Du, D.H.C. (1995). Issues and challenges in ATM networks. Communications of the ACM, 38(2), 28 29. Weiss, J. (1990). LAN/WAN internetworking. Telecommunications, 24(7), 57 59.

Source: Network World, Oct. 30, 1995, p. 62.

Supplement 6.3: Projected pricing of ATM
The price of ATM access is predicted to drop consistently at least for the next three years. The predicted drop is as follows: 1995 1996 1997 1998 $5400 $4000 $3000 $2000

Sources: CIMI Corp., Voorhees, N.J., US; printed in Computerworld, Nov. 20, 1995, p. 2.

Alexander, P. (1995). Network management: the road to ATM deployment. Telecommunications, 29(9), 47 50. Basi, J.S. (1990) Networks of the future. Telecommunications, 24(7), 33 36.


. . .the intrinsic value of a telecommunications system grows combinatorially with the number of subscribers it interconnects. David Rand Irvin

ISDN is the acronym for Integrated Services Digital Networks. It is a digital version of the switched circuit analogue telephone system. ‘But,’ say the critics, ‘we have had switched networks and telephones for a long time. So why the excitement?’ ‘Well,’ reply the proponents of ISDN, ‘it is multimedia and handles data, voice, images and video.’ ‘But,’ add the sceptics, ‘we have had voice and fax through the modem all these years, and so what is new?’ The proponents of ISDN then point out that it is not only a technology for multimedia communication, but it is an enabling technology that will allow computers with applications of integrated multimedia to be as ubiquitous as the telephone is today. The sceptics then say, ‘I have been hearing of the ISDN for over a decade and we have seen no results. Maybe we should not upset what we already have. We may not need something that is so complex and difficult to implement.’ ‘Yes,’ counter the proponents of ISDN, ‘It has taken a long time and will take longer because we are dealing with a system that is not only integrated but internationally so. And getting international agreement between the carriers, suppliers of telecommunication components, the computer industry, and many governments, does take a long time. It also takes a long time for testing a product carefully and then getting acceptance especially for an advanced concept. Remember, there were only 25% of households with telephones in the US in 1920 and it took 60 years for this percentage to increase to 96%. ‘These things take a long time’. So the argument rages. In this chapter we will examine the myths and realities of ISDN. We will describe ISDN as it is

today, look at its objectives for tomorrow and the day after, discuss the implementation of ISDN, and examine the obstacles and future for ISDN. We start with image processing followed by voice processing. In both cases, we examine the nature of the application and their uses in business and daily life. These applications are constrained by a lack of resources necessary for implementation of the digital technology. These constraints are then examined. For the reader who wishes to read on, there is more on the nature of ISDN and its evolution.

The computing environment
ISDN was designed for an environment where all the needs of computing could be integrated. This included applications of data and voice, as well as images and video. Data and images can be easily digitized while voice and video are basically analogue signals and more appropriate for telephony than for a digital computer. The problem is one of integrating the two types of signals or else pay for the inefficiencies resulting from the interfacing of the two. Either all should be digitized or all be analogued. ISDN takes the approach of all being digitized. However, understanding the nature and magnitude of the digitized and analogue applications is necessary to appreciate ISDN. All the early processing were computations and transaction processing which were numerical and digital. Later on we added textual processing, but this was digitized by giving each alpha character a unique digital equivalent. Then came graphics which in many cases could be digitized if a curve is viewed as a set of lines, which they are because many discontinuous lines can look continuous. 69

Telecommunications and networks Likewise an image (like a drawing or even a photo) can be viewed as a set of dots where each dot is digitized. Thus, an image can be represented by an array of numbers. A number code could also represent the intensity (and perhaps colour, if we are not just dealing with black-and-white images). These primitive picture elements are called pixels. Thus a computer image is a two dimensional array of numbers, the individual pixel values. For example, we might have a 100 ð 100 array of intensity measurements, each selected from a range of 0 to 100, where 100 represents white and 0 represents black, and the 99 intermediate values represent various shades of grey. The initial stage of image processing is pixel processing and may involve a ‘clean-up’ process, i.e. removal of noise (e.g. black pixels that should be white) that is often introduced by the hardware that generates the pixel image. A ‘smoothing’ operation is then performed in which a small cluster of adjacent pixels are compared, and a single odd-valued one is adjusted to a value that is closer to the ‘average’ of that of its neighbours. This smoothing operation is computationally trivial but done repeatedly, something like 10 000 times on a 100 ð 100 array of pixels. One can quickly see that the computational needs of image processing can multiply rapidly and become very large very quickly. Digital computers soon became indispensable. The large set of computations for image processing is worth the price because as the saying goes ‘a picture is worth a thousand words’. Also, there are many applications in business that range from simple charting for a report to complicated drawings, blueprints, and CAD/CADD (computer aided design/computer aided drafting design). In medicine, magnetic resonance imaging (MRI) and computerized axial tomography (CAT) scans of parts of the body including the brain may well avoid dangerous surgery and save many lives. In a less dangerous and more entertaining way, images are used in cinema and film-making. In 1982, the film Tron, was released and credited as the first feature film that used computer generated imagery as background for live actors. Well, within a decade, this digital technology became so commonplace that it is now being used throughout the entire film industry. We now have digital cameras that take moving and animated pictures and store them for future manipulation by computer. This technology 70 has applications in business for training, for advertising products and for simulations in decision-making. Image processing is also important in any office. To give you some idea of the magnitudes involved, consider a study done by Arthur D. Little in 1980 in the US. The study found that an average office worker handles per day: 1 page from files, 5 pages from mail, 4 from catalogues, 11 photocopies, 32 pages of computer printout as input, 14 pages passed along, 5 pages mailed and 8 pages to be filed. Now project this into the future and you can soon see that there is a great potential demand for image processing if office work is to be rationalized and made efficient, especially when office processing shifts from data and text only to their being embedded in graphics and pictures as images. Even the need for processing data is increasing rapidly as shown in Figure 7.1 where files in the 1970s have grown into very large files in the 1990s and may now occupy 1012 bytes. In addition to data, text and graphics, there is the need to process voice which is often indispensable today in the office, in the factory, in the home and in all walks of life. This need is largely met by the telephone and partly by voice processing. The differences between the processing of voice and that of data may be well known, but for the record they are summarized in Table 7.1. Many applications related to voice are processed by the telephone exchange. An example is video processing. One configuration is shown in Figure 7.2. This can be done through a central facility like a PBX (Private Branch Exchange) carrier with a BRI (the basic rate interface) rate for low volume, or through the LEC (Local Exchange Carrier) paying a PRI (the primary rate interface) for high volume. The BRI and the PRI are two classes of services to customers of baseband ISDN. The BRI provides up to 144 kbps (two 64 kbps ‘B’ channels C one 64 kbps ‘D’ channel for control information). The PRI provides up to 1.54 Mbps which includes twenty-three 64 kbps ‘B’ channels and one 64 kbps ‘D’ channel. The many applications of voice and image processing mentioned above are stand-alone applications. Many can be integrated providing a raison d’ˆtre for ISDN. Some of these applications e are listed in Table 7.2. They vary in bandwidth demand as shown for a sample of applications in Figure 7.3.

Bytes of data 1012 1011 1010 109 108 107 106 105 104 103 102 101 Large file Small file Blocks of records Large record Small record Very large database Integrated database Large database Medium database Small database Very, Very large database




Figure 7.1 Growth of corporate databases


PRI Inter LEC PRI Exchange



BRI = Basic rate interface; PRI = Prime rate interface; LEC = Local exchange carrier

Figure 7.2 Networking with video-telephones


molecular models. 1993: p. large data repositories interconnected for the purpose of processsing medical images. a facsimile page that took 30 seconds in the pre-ISDN era would now take around 4 seconds. The answer is that modems are appropriate for PCs with a 9. appropriate for either data or voice transmission. speeds of the B-ISDN can get data rates between 10 and 600 Mbps. The most spectacular application of ISDN (not even listed in Table 7. for applications beyond 100 kbps. Applications of ISDN were listed in Table 7. voice network Computer network Voice network Voice Long Limited Public usually Service charge Usage: Distance: Performance: Ownership: Charges: Digital Limited High Private Low.6 kbps modem. only approximate Table 7. and the like are waiting adoption. which is a 16 kbps designed to control . one needs the B-ISDN (Broadband ISDN) as shown in Figure 7.2 because it is still a proposition) may well be NASA’s Earth Observing System for global-change research. Not to scale.1 Computer network vs. The B-ISDN is designed for voice and video as well as large volumes of data to be transmitted over 72 long distances. Such bandwidth applications like the high speed workstations.3 Bandwidth requirements. What is ISDN? Voice and data are inputs to an interface hardware equipment which is connected to an ISDN interface. if any The resource environment The question that could be asked is why these applications cannot be processed by our conventional digital equipment along with modems to do the conversion between digital and analogue signals. despite their steady increase in bandwidth from around 100 bps to almost 50 kbps.Telecommunications and networks Telex Desktop teleconferencing Text Visualization File Transfer FAX CAD/CAM Digital videophone Transactions Analogue phone 10 100 1K 10K 100K 64K 1M 10M 100M 1G Bandwidth (bps) Figure 7. distributed CAD/CAM (computer aided design/computer aided manufacturing). we are talking about B-ISDN without really explaining the baseband ISDN or narrow band ISDN. 43). Up to 100 kbps. and one ‘D’ channel. However. That is to begin in the late 1990s (Irvin. It is connected to an ISDN switch through three channels: two ‘B’ channels (bearer channels) of 64 kbps and 16 kbps. However. For example. It is time to do so. With fibre optics.2. which is expected to transmit more than one trillion bytes of data per day (or equivalently 92 million bits per second) for the duration of a 15 year period.4. the baseband ISDN may well be adequate for connecting LANs as well as for transmitting data faster than before.

The implementation was ahead of schedule and more so in Europe than in the US. switched or packet switched networks. reset calls and receive information on the incoming calls including the identity of the caller). implemented and tested. This is shown in Figure 7. ISDN was implemented following all the rules of good development: the system specifications were stated by the users.2 Applications made feasible by ISDN Customer sharing with salesperson the same screen on products and conditions of sale Teacher sharing screen with student Telecommuter sharing screens of multimedia desktop with customer. the development had to be global and this takes a long time. 1990: p. This may be because in Europe we have nationalized PT&T 73 transmission in the ‘B’ channel (used to signal the switching system to generate calls.g. Implementation of ISDN ISDN has no system to compete with or to match and copy.4 Rise in data rates (adapted from Griffiths. supervisor or co-worker Video-conferencing with all parties looking at data. based on experiences with the Red Book. In 1984. that is. pictures and even results of a simulation in progress Access and dialogue with librarian Medical records and imaging access Remote medical diagnoses like from an airport or form Security (and identification) and surveillance system Teleshopping Telebanking Telereservations Telenews High speed bulk multimedia transfer including books from a library Multidocument image storage and retrieval Note: All the above applications involve on-line remote processing. Suppliers for components were selected: Siemens Stomberg-Carlson in Europe and AT&T and Northern Telecom in the US. and the system was designed. Many of the above (e. Similarly. 158) Table 7.ISDN Bits/second ? 10M B-ISDN 100K ISDN 10K MODEM 100 1960 1970 1980 1990 2000 201 0 Figure 7. graphs. These three channels are sometimes referred to as the . text. the first specifications of ISDN appeared as the ‘Red Book’ specs followed by the ‘Blue Book’ specs. an international organization telecommunications based in Europe. These channels connect to the ISDN switch which is connected at the other end to networks that may be signalling. Since the system was to have a world-wide relevance. one 64 kbps can be submultiplexed into two 32 kbps or eight 8 kbps channels for eight terminals connected in parallel. They can be multiplexed to form one 128 kbps or by multiplexing four ‘B’ channels to form one 256 kbps channel. non-switched. Two of these three channels are of 64 kbps each. These were made by CCITT. video-conferencing and those involving dialogues with up to the minute updated data) are isochronous.5. they are time dependent and in real time 2BCD system.

US companies are developing alliances and partnership to redirect their resources to the burgeoning potential market of ISDN. there were over forty different trials of services of the ‘Red Book’ standards that were in progress. adapters. 1993: p. 50). 1994: p. (Lai et al. painful and expensive though it was bound to be. These are: national ISDN not available. and a schedule for providing such service to all subscribers by the year 2015. 49) 17% of the companies surveyed rejected ISDN. lack of standards for ISDN. the infrastructure to be required by ISDN was being laid. 36).) PC or workstation ISDN interface ISDN switch Switched Network Figure 7. 1993: p. nation-wide ISDN not available. By 1988.. not able to justify costs. the availability of ISDN has been 100% for several years. in Mannheim and Stuttgart. services are expected to be around 90% by 1996 (Galvin and Hauf. while in the US the telecommunications industry is decentralized. et al. and in Brittany. In Australia there is also a 95% availability of ISDN. the use of fibre grew from 456 000 miles per year to 3 811 000 miles a year by the end of 1991. not compatible with organization’s computing environment. unattractive tariff structure. not an established technology. The same study identified the principal obstacles for the adoption of ISDN (Lai. not available in our area. Germany in 1987. In the US. In the US. The private carriers had billions of dollars invested in analogue equipment and so there was hesitation and caution but no resistance to a conversion. The main reasons cited (in order of significance) were: ž ž ž ž ž ž ž other networks can serve the same communications needs equally well. In a study done in the US. available only in metropolitan areas. The implementation of ISDN was started in the UK in 1985 by British Telecom followed by Illinois and California in the US in 1986. But carriers in the US did anticipate the coming of ISDN and have rapidly added WAN access capabilities to everything ranging from computers to internetworking equipment to video-conferencing equipment. ISDN services not attractive. world-wide ISDN not available. also in 1987.Telecommunications and networks Switched Network B (could be data) B (could be voice) D (could be info. incompatible equipment. Japan enjoyed a fibre optic backbone network since 1989. Despite its extensive testing and its international certification. expense of ISDN equipment. the Euro-ISDN (European ISDN standard) was introduced in 1988 and compliance 74 with the standard was targeted for 95% in 1995. In Europe.5 Basic components of an ISDN (Post Telephone and Telegraph) organizations. international ISDN is not commonly available. bridges and bridge/routers (Garris. The computer industry is also producing equipment necessary for an ISDN environment such as the ISDN modems. France. In many countries. .. lack of user awareness. 1995). ISDN has not been widely adopted. They announced their fibre-to-thehome trials. In the UK.

This may well come without any grand opening and fuss but will just creep upon us like a long Table 7. These technologies are compared in Table 7.ISDN Summary and conclusions ISDN is a viable technology but has not yet received wide acceptance. Meanwhile it will coexist with other technologies that are often older and better entrenched like the analogue. it is a spin-off of ISDN (Bhushan.3. voice and video into one single digital end-to-end system of seamless transmission and communication. Such integration will not only enhance the many applications of integrated data digital and voice as listed in Table 7. The integration of the two basic media of digital data (from the computer) and analogue signals (from a telephone) are shown in Figure 7. 1990). Switched 56 and SMDS. Another competing technology is the frame relay.2 but will encourage the user and the industry to identify and develop applications that have only been dreamed of. The ‘I’ in ISDN stands for the integration of the media of data. This is not older but younger than ISDN.3 Comparison of ISDN with other technologies Advantages Disadvantages Slow Long set-up times High monthly services charge Confined to local carrier’s region Too expensive for low frequency users Not universally available or compatible Unfamiliar to many professionals Expensive if packaged wrongly Analogue: Switched 56: SMDS: Frame relay ISDN Ubiquitous Cheap Faster than analogue Provides digital service Highest bandwidth Best for full time connections such as large and very busy branch offices Fast Flexible configuration Can be cheap PC Telephone ISDN LAN bridge National ISDN Network Multipoint ISDN Bridge or Router Printer Host computer Office computer Peripheral Figure 7.6.6 Integration of digital and analogue 75 . it may well be possible soon to pick up the phone and call someone across the continent or even across the world and download a database or images or video in one window or the screen and see the other party in another window simultaneously. in fact. For example. images.

ISDN is a foundation technology and like the foundation of a house it is fundamental and even invisible. With compression. not even commonly known. There is still confusion and lack of awareness about availability schedules. ISDN in France 1987 1990: from the first commercial offering to the national coverage of numeris ISDN. A broadband ISDN. as well as interoperability and portability between components and devices using ISDN. Francois Lecrec.2: ISDN in France The Numeris ISDN in France has many applications using ISDN that can be classified into four categories as shown below. ISDN is not yet a plug-and-play technology. That gave an order of magnitude slower than Ethernet but an order of magnitude faster than an analogue modem. The alternative was to get a T1 link for $700 per month plus a bridge and a CSU/DSU (channel service unit/data service unit) for $12 000. it carries a bigger load than does the asynchronous traffic. In 1990. the FDDI backbone using Ethernet bridges. Dec. Roger Trulent and Pierre Deffin. January 1991. standards. the problem of cheap and stable rate structures. but it is a big leap from the analogue-only communication of the telephone. and even the benefits and enhancements of ISDN. ISDN is an access technology that allows us access to all data whether in the form of digits or analogue signals simultaneously. All this will come in due time and as with the telephone. The planning and definition of this architecture requires two to three years before becoming available. it may take many decades but it will be ubiquitous and indispensable for a worthy style of living that awaits us. This cost $40 per month for each line and $2100 for a bridge that lashed the two 64 kbps ‘B’ channels of ISDN to create a 128 kbps pipe. rates. is now being implemented (a case on the development of its standards is discussed in the chapter on standards). ISDN is constrained by bandwidth. 30 35. IEEE Communication Magazine. Also. the packet switched network. There are still problems of national standards being in conformity with international standards. which are leased lines. the B-ISDN. 1993. the second stage was the voice and data integration. Case 7. . p. which is the subject that we will examine in some detail in our next chapter. It is not commonly accepted. ‘The various needs expressed by a corporate network are split amongst the three main bearer networks. But the infrastructure for the house of ISDN has been carefully planned and well implemented. 75. Also being synchronous. It is a vehicle by which business as well as households will cruise on the international information highway. and the third stage was the full integration with the architecture of the corporate network.1: ISDN at West Virginia University (WVU) WVU had a FDDI backbone that links its ten buildings but had 90 other buildings that could not be interconnected. Economic analysis indicated low bandwidth connectivity in these satellite buildings dial-up routers that can use modems to link LANs over ordinary phone lines gave only the most casual support of LAN interconnection despite compression being used. ISDN is consistent with many of the well accepted architectures of networking.Telecommunications and networks overdue progression of computing applications. Therese Morin. The first stage was to use ISDN for a specific application including image videotext. pp. Source: Byte. thruput exceeded 200 kbps. We have not yet reached the level of applications that provide an industry with the assurance of economies of scale for the industry to adopt ISDN. DATA Remote processing LAN interconnection Software loading IMAGE PROCESSING Image server Medical Imagery Telesurveillance Remote teaching Video Telephony Local image stations MULTIMEDIA Radio commentaries Teleconferencing Audiogram Service Illustrated video text DOCUMENTATION High speed facsimile Electronic mail Document database Document exchange Case 7.’ Source: Jean-Pierre Guenin. WVU decided on a flat-rate ISDN service that connected the satellite buildings to 76 The ISDN implementation was done in three stages. and the ISDN.

(1994). the players are able to communicate with each other and the Americans nudged the Norwegian to speed up. Derfler F. the author was playing competitive bridge in real time on the Internet. Telecommunications. 39 42. ISDN sleight of hand. Telecommunications. 39 42. C. 7(1).E. Telecommunications. (1995).M. (1994). ISDN solutions: ready for prime time. . ISDN explained. ISDN personal video. and Dent. B. 72(3)... J.R. Carr.S. Crouch.’ ‘Oh’ I said. H. 13(18). ISDN may be here to stay . Garris. Three of the players were from America and one.. B. J. S. AT& T Technical Journal. But it’s not plug-and-play. J. Guynes.M. Viola. and Hauf. Making broadband ISDN successful. And what does it cost in America?’ asked Atle.3: ISDN for competitive bridge across the Atlantic In December 1996. The American from California responded: ‘I do not need an ISDN. Next month I am getting an ISDN connection. 28(10). (1994). with the screen name Atle. J. was from Norway. W. Irvin.J. Lai. but soon got immersed in playing bridge and the conversation on ISDN ended. NE1 6. Galvin. PC Magazine. 10(4).’ ‘Wow’ said Atle. 33 38. In between the plays. Jr. Wiley. (1993). 40 45. D. 46 52. Walters. Bibliography Bhushan.D. P. Atle was consistently slow in his responses. 29(6). Expanding the market for ISDN access.J. 77 . (1990). Hicks. V. IEEE Communication Magazine. (1994). . I get a fast enough response and unlimited access for less than $20 per month. Sankar. 28(10). and Jetzt. 35 38. (1994). A new direction for broadband ISDN.S. IEEE Network. A. A. 24(7). Information Systems Management. J. ‘How much will that cost you?’ $US500.J.A. Finally Atle responded: ‘I apologise for being slow but not for long. (1993). A user’s guide to frame relay. 167 187. and Bordoloi. (1990). 55 57. 14(5). 27 33.L. ISDN: adoption and diffusion issues. Telecommunications. PC Magazine. Griffiths. M. Betting on the dream.ISDN Case 7. (1991).

in the case of networking. Finally. we make the exception on pedagogical grounds. Across the Atlantic. including DNA. There were many others developed in America. In 78 this chapter we will examine each. We conclude with observations on how this rivalry between the three models will affect the end-user as well as the computer industry and telecommunications in general. Systems network architecture (SNA) SNA was developed by IBM beginning in 1972. The SNA architecture consists of sets of layers within groups of the physical unit (PU) and the logical unit (LU). The architecture of network systems is the style and design of the structure of networks that enables electronic communication. It is an overview of the structure of networking. we will compare the OSI models with APPN and TCP/IP. We will then look at APPN. each layer provides a set of functions so that communication can take place not only for file transfer and transactional processing. there was IPA and XBM developed by ICL in the UK. It is a layered architecture where each layer is a group of services that is complete from a conceptual point of view. but for the management of on-line and real-time dialogues that may take place between two parties. as well as the OSI model developed by the international organization CCITT. devices and protocols involved. The earliest architecture of networks was SNA by IBM. Thus. SNA has a lot of clout. The traffic expected was not only data but also voice and video. And there was a shakedown as expected. but using multiple protocols is the trend Anon. and style and design. DCA. the updated version of SNA. Each layer has one or more entities. But. the TCP/IP. First. 1995 Introduction Architecture is concerned with structure. layer 1 is the physical layer that provides the mechanical and electrical level interconnections for the two stations (sending and receiving).8 NETWORK SYSTEMS ARCHITECTURE OSI is a lovely dream. It is like travelling around much of the world and then looking at the world map. It provides a framework for all the components and protocols discussed in earlier chapters and provides a frame of reference for much of the discussion to follow. but three models seem to have emerged. Typically. It was expected that all these models would fold up or merge into one internationally accepted standard. where an entity is an active element within a layer. Each entity provides services to entities in the layer above them. In addition. and in turn receive services from the layer below them (except for layer 1). Networks and telecommunications are very rich in the names and acronyms of the many components. An early overview of network systems architecture could be very intimidating. Using them in an overview without explaining them first could be difficult for both the reader and the author. we will examine SNA and OSI and compare them. and now we are ready to see how they all fit together and provide a basis for the remaining discussion. TCP/IP and OSI. OSA and TCP/IP. Layer 2 is the datalink layer that provides . and compare it with its competitor in the US. Each layer is numbered sequentially from 1 to 7 starting from the bottom. SNA. We have done the opposite. an overview is done early in any text in order to facilitate placing the components in their relative position. TCP/IP is current reality. So we discussed important components and protocols.

This layer is sometimes called the path control layer because it is responsible for the path that a message takes through the network which could include more than one node. and controls the mode of sending. access rights. This layer controls the pacing of data within a session. SNA assumes the use of different approaches including national and international standards already in place. there is the upper most layer.1. Both paths OA and BD are controlled by network software which is additional to the OS (operating system) software END-USER with applications LOGICAL UNIT LAYER 7 LAYER 6 LAYER 5 LAYER 4 LAYER 3 LAYER 2 APPLICATIONS PRESENTATION DATA FLOW CONTROL TRANSMISSION CONTROL PATH CONTROL DATA LINK CONTROL PHYSICAL CONTROL PHYSICAL UNIT LAYER 1 Figure 8.1 Seven layers of SNA 79 . the LU2 supports sessions with a single display terminal of the 3270 type while LU6 supports peer-to-peer connection with the application subsystem or application program. This network layer 3 is serviced through routers. The PU. and error detection and correction. there is layer 4 which is for transmission and provides functions for error-free delivery of messages such as the flow control. LU. layer 7. arbitrates users’ rights and services when there is a conflict. In the logical unit. through path BD. Layer 5 is for data flow control. It initiates and establishes connections and keeps track of the status of sessions and connections. but SNA does not actually define specifications for protocols in this layer. This includes the control of exchanged information. end-user and the seven layers of network architecture are portrayed in Figure 8. and document distribution and interchange. For communicating in the SNA schema. receiving and response. Instead. database management.2. the application layer. Layer 3 is the network layer that provides routing of messages between two transport entities. error recovery and acknowledgment. This layer also provides an optional data encryption/decryption facility (to be discussed in a later chapter).Network systems architecture rules for transmission on the physical medium such as packet formats. layer 7 as shown in Figure 8. The message then goes down to the physical unit (path OA origin to the point A). file transfer. The physical layer in SNA is addressed in the architecture. It is responsible for the transmission between two nodes over a physical link and is serviced through bridges. and finally up the seven layers to the destination D. The sessions’ medium will vary in mode and hence there are seven LUs that correspond to each type of session mode. the message must first go to the top-most layer (closest to the application and end-user). The LU (logical unit) is between the PU (physical unit) and the applications of the end-user who enters into sessions for communication. It provides the application service elements for the end-user. and resource sharing. Finally. across the transmission path AB. Layer 6 is the presentation layer which provides the necessary services for formatting different data formats used in each session and manages the sessions dialogues. synchronizes data transfers. For example. operator control over sessions.

terminal devices. The communication path is shown in Figure 8. to provide an architectural reference point for network design. the IEEE 802 standards of the Institute of Electrical and Electronics Engineers in the USA developed protocols for the physical layer as well as for medium access control in the datalink layer and for logic control for most of layer 3. The transmission is through the network. Thus. The IEEE 802 included standards and definitions that became de facto standards for the industry for the Ethernet and token rings. Other protocols developed were for other sets of layers such as the IP for layer 3 and the TCP for layers 4 and 5. The interconnection refers to procedures for exchange of information between computers. to serve as a common framework for protocols and services consistent with the OSI model.2 Communication through the layers and the applications software. a subject that we will now address.3 Early sources of protocols 80 . and SNA Layer Layer Layer Layer Layer Layer 1 2 5 4 3 2 IEEE 802 DOD CCITT ISO Layer 1 Logic control Medium access control Physical TCP TCP IP Sessions TP X-25 LAP-B X-21 Figure 8. people. The OSI model OSI stands for the Open Systems Interconnection reference model by the ISO. The main objectives of the OSI model were: 1. These protocols were developed by the DOD (Department of Defense) in the US. there were protocols being developed that addressed the individual or group of layers as shown in Figure 8. Corresponding to the use and development of the SNA layers. International Standards Organization. 2. processes and networks. as well as providing definitions for concepts like connection and connectionless service.2. The protocols by the CCITT and the ISO corresponded directly to the OSI model.Telecommunications and networks O Origin (Application) LAYER 7 LAYER 6 LAYER 5 LAYER 4 LAYER 3 LAYER 2 LAYER 1 A 3 2 1 Data / Voice / Video Transmission Network 3 2 1 (Application) Destination D LAYER 7 LAYER 6 LAYER 5 LAYER 4 LAYER 3 LAYER 2 LAYER 1 B Figure 8. The open refers to a specification made openly in the public domain in order to encourage third-party vendors to develop addon products to it. OSI was announced in 1977 as a response to the need of international standards for communications and networking. We mentioned them in an earlier chapter and will come back to them later in this chapter. Other early protocols were developed by the international organizations like CCITT for layers 1 3 and by the ISO for layers 4 and 5.3.

there is no direct counterpart in the SNA model for the lowest layer of the OSI model..4. however. The OSI model never did catch on in the USA despite the fact that it was developed by an international organization. 1995) Source: When initiated: Architectural design: Levels implemented in 1985: Acceptance: Future: IBM 1972 74 De facto standard in the US and IBM mainframe users All seven layers Welcomed by IBM mainframe users and many in the US Updated by APPN and other products being steadily introduced 81 . The European Common Market Commission had made the OSI and ISO its standards for connecting systems products and networks of different OSI 7 APPLICATIONS In design. except at the lowest level and physical layer. While there is an almost one-to-one mapping for most of the upper layers. was (and is) very popular in Europe.1. OSI. to facilitate the offerings of interoperable multivendor services and products. For a summary comparison on SNA and OSI in general terms see Table 8.Network systems architecture 3.1 A comparison of SNA and OSI SNA OSI ISO (International Standards Organization) 1977 Largely like SNA but has more for the physical layer 1 Bottom three layers Largely in Europe with SNA/OSI interfaces negotiated with IBM Work being done at all seven levels in addition to the network management level (see Deirtsbacher et al. as is clear from the conceptual comparison in Figure 8. communications devices outside its model: the OSI did not have any such reservations. the OSI model was not very different from the SNA model.4 SNA versus OSI Table 8. The SNA model left the definition of the SNA 7 APPLICATIONS 6 5 4 3 PRESENTATION DATA FLOW CONTROL TRANSMISSION CONTROL PATH CONTROL 6 5 4 3 PRESENTATION SESSIONS TRANSPORT NETWORK 2 DATA LINK CONTROL 2 DATA LINK 1 PHYSICAL CONTROL Externally defined 1 PHYSICAL LEVEL Figure 8.

This was appropriate for the 1970s when SNA was first implemented. APPN now kept track of network topology making it considerably easier to connect and reconfigure nodes. and computing resources without host interventions. 1990: pp.Telecommunications and networks manufacturers. In the USA. The routing was now possible without host intervention. In addition. bridges. 1986. IBM was no small firm to be pushed around. IBM not only had a dominance in all segments of the computing market but also a sales presence in all the world. and the third is to have an SNA/OSI interface in between (Tillman and Yen. however. In this chapter we will examine how IBM responded to the changing environment with its APPN and the competition it faced from TCP/IP. Such enhancements to the APPN added more functionality and made the APPN more robust (a program that works properly under all normal but not all abnormal conditions). p. both sides have equal responsibility for initiating a session. 1985. In 1985. IBM employed 405 535 people (Datamation. The hierarchical host to terminal configuration of the SNA is now replaced by networking that handles peer communications among hosts. But IBM did not give in to the OSI model. It seemed that the model was bound to become the international standard. bridge/routers. but the world of computing changed in the 1980s.) In 1985. performance increased. June 15. IBM continued with its proprietary network architecture and had a ‘captured’ market from all the users of IBM mainframe computers. APPN is the acronym for Advanced Peer-toPeer Networking. Since then there have been many network announcements to enhance the APPN.554 billion. An important enhancement was the SAA. Also. 219 220). IBM had all its seven layers implemented while OSI had only three bottom layers implemented. there still is strong competition from TCP/IP. and is compatible with the hierarchical SNA structure while maintaining connectivity with dumb terminals. IBM had a revenue of $48. 56). The ten messages required under SNA now required only two messages. centralized management was replaced by decentralization of computing. But despite all the enhancements to APPN. it signifies the headers for transfer from a high protocol level to a lower protocol level). more than all the next 12 world-wide competitors combined. the Systems Applications Architecture. Internet transmission. There are many interconnection devices that include repeaters. Because APPN is an update of SNA. 1990: pp. the new traffic can still be handled without much conversion or encapsulation (in telecommunications. This is in contrast to the master slave relationship. including Europe. This new computing paradigm was for peer-to-peer computing and client server computing. the DOD (sponsor of TCP/IP) had executive directives to use international protocols. where only the master unit initiates and the slave responds. 216 218). and interstation transmission (Tillman and Yen. One was that at the time of the most pressure. We shall discuss these configurations and the downloading and downsizing of the mainframe to the enduser client and the server in a later chapter. There are many reasons for IBM’s hesitation to give up SNA for international harmony. The Europeans used its leverage as a large customer of IBM equipment to persuade IBM to agree to the OSI model. APPN was originally targeted in 1986 of midrange systems. that had a sophisticated and fundamental routing technology. The dumb terminals were replaced by PCs and workstations. allocation of resources by the host was to be replaced by the sharing of information 82 . programs and peripheral services. It did. departmental computers and desktop PCs. routers. In peerto-peer communication. There are three ways to integrate the two models: one is directly at each of the seven levels. (This revenue was more than the GNP of all the countries in Africa outside South Africa and two of the oil rich countries of North Africa. the second is indirectly through the three bottom levels of the physical unit. gateways. device emulation. The APPN The problem facing IBM was that its networking architecture in SNA was designed for a mainframe host serving a lot of dumb terminals in a hierarchical master slave configuration. agree to a gateway between the SNA and the OSI model. the mainframe host was being replaced by computers that were more powerful than many mainframes as servers of computing resources including databases. and the single host was now being replaced by multiple hosts and servers. A peer is a functional unit on the same protocol level as another.

1993: p. TCP/IP took on a life of its own. The routing was adaptive giving the ability to route around failed circuits. The access network is within the sphere of influence and control of the end-user Intelligent terminal PC Access router Access router Dumb terminal Workstation File server Access device Backbone device Figure 8. The TCP/IP was expected to retire. perhaps close to 20% of the over 50 000 SNA networks across the world serving some 300 000 nodes (Kerr. The other approach is to implement SNA gateways on UNIX machines. sharing responsibility for the different layers in the network architecture with SNA still specializing in the upper layers. Incidentally. We visited TCP/IP earlier as a set of protocols. unable to resist the pressure from IBM’S SNA and OSI supported by the international community.Network systems architecture TCP/IP TCP/IP stands for transmission control protocol over an Internet protocol. The two are independent enterprise architectures but they can coexist on the same physical network. even IBM approved of the SNA/IP alternative to internetwork. They were developed by the DOD in the mid-1980s and was concerned with the middle layers of the seven layer cake architecture. or both. it was generally accepted that TCP/IP had made many inroads into the SNA market. as LANs expanded. This allows enterprises to deploy pure APPN or IP backbones. 28). One way is to run TCP/IP on a IBM mainframe using SNA. This enables IP routers to become the key to controlling broadcast storms and ensure reliability in large meshed internet backbones connecting local and remote LAN users’. The backbone is connected to access networks through access devices. There are several vendors that offer routers that handle both IP and APPN simultaneously. the routers were enhanced to accommodate more protocols and to provide critical segmentation capabilities. Since it was based on the UNIX operating system. as shown in Figure 8. it had the support of many UNIX users who liked the multitasking and multithreading features which are ideal for multiplatform networks. for IBM offered the system as an alternative to their pure APPN. It is very tempting to marry the SNA and the TCP/IP. (Rosen and Fromme.5 Access and backbone networks 83 . 79). UNIX was becoming the system of choice for many mission critical applications in IS (information sytems). 1992: p. What is emerging though is the use of TCP/IP to create what is called the SNA/IP networks that competed directly with the APPN. The success of the TCP/IP approach was that it relied for its critical backbone on the multiple protocol routers and the OSPF: the ‘open shortest path first’ approach that was a new routing algorithm. But.5. In the early 1990s. ‘These routers evolved out of the TCP/IP and Ethernet Communications processor PC environments.

The X designation tells us that it has been developed by CCITT. The backbone may be a MAN or a WAN with wide-area infrastructure for communications across the region or country. As defined by the IEEE 802. but the dream of the network manager may well be of having the ability to mix and match not just topologies and protocols but also network management strategies. Another view. In APPN networks. Achieving a homogeneous system may not be as easy as one would like. The alternative strategies available to the network manager are the subject of the next part of this book.7 where there is often a choice in each layer. Multiple protocols give flexibility to the network manager.6. or LAN-attached communication controllers. start and end delimiters. One or more token rings could be used with an IP backbone. These access devices channel traffic to the backbone routers where the data fields for the backbone include addresses of source and destination. linking to all points of the enterprise network. Multiple protocols There are many protocols under almost continuous development. This is shown in Figure 8. communication controllers.8. FEPs (Front End Processors). There are two routers accessing the backbone. the access devices can be routers. This view is shown in Figure 8.400 to X.5 standard.6 Token rings and backbones 84 .430 defines standards for a general purpose system and recommendations that may solve many of the interconnectivity problems like those of e-mail and EDI (Electronic Data Exchange) so important for financial institutions. start and end delimiters. In the absence of internationally agreed architectural model and protocols we must mix protocols within the seven-layer architectural model. or even globally across the world. The access devices are routers or communications processors. is the use of the OSI model and all its protocols. the token ring frame contains fields for the addresses of source and destination. terminal (dumb or intelligent). and frame check sequence. and fileservers. and frame check. PC or workstation Token ring Routing bridge Token ring Access router BACKBONE Access router Token ring PC or workstation Figure 8. that from the other side of the Atlantic. This is shown in Figure 8. where there are two token rings on the sender’s side with a source routing bridge in between and one token ring on the receiver’s side. workstations. The set of recommendations from X.Telecommunications and networks and could include PCs.400. like the X. One way to link SNA terminal traffic and a LAN is to put a token ring LAN adapter in the SNA device.

FTP (File Transfer Protocol). in this chapter. and TCP/IP protocols.B. CCITT X. They are the basis of hardware and software products that are interoperable.5 (Token Ring). Frame relay.721-725 ISUP CCITT Q. CCITT X. and Other processing Applications Application Support & Interface Transport Interface Transport OSI. Cell/Packet Figure 8.9. X.24 CCITT X. q. In this chapter we have taken an overview of some of the models. SCCP CCITT Q. CCITT X. TCP/IP Data Link IEEEE 802. CCITT X. Network architecture is a complex subject. (Ethernet) IEEE 802. We have discussed access methods in a previous chapter.10. TCP/IP and OSI are shown in Figure 8.225 ISO 8073.771-776 6: Presentation ISO 8823.25 Figure 8.e. SNMP (Simple Network Management Protocol). A summary comparison of APPN. SNA.400. The main reason is that ISDN had no competition. In 85 . with many books being written on each of the main models (see Meijer and Peeters. Martin. & others Common Transport Semantics SNA. The different architectural models being accessed through a gateway network are shown in Figure 8. software. the international OSI model. 1987. OSI. X. a topic which will be discussed in a later chapter.25.761-766 5: Sessions 4: Transport 3: Network 2: Data Link 1: Physical ISO 8327.7 Protocols at different layers 7: Application ISO 8650. 701-707. FDDI.21 Bis.21.711-714 LAP.Network systems architecture Transaction processing. 9072. X. access methods and protocols.226 TUP CCITT Q. we discuss protocols.25. APPN.22x TCAP/CCITT Q. 1988. X. There is no general acceptance for a single model for network architecture like there is one international standard for ISDN.701-707 CCITT Q 701-707. Message processing. Q.3.8 International protocols Summary and conclusions Network architecture is the design of a communication system including hardware. Meijer. We examined the SNA along with its updated APPN. CCITT X. Database processing. and Tang 1992 for but a small sample). TCP/IP i. 1982.

logically integrating their applications. OSI (adapted from Friedman. integrating vertically with their DSSs and EISs and ESs (expert systems). in the mid-1980s. they were actively looking for network performance and a network architecture that worked. TCP/IP vs. They were preoccupied with expanding their applications portfolio.9 Appn vs. while those with UNIX systems preferred the TCP/IP partly because these protocols had a built-in integration with their operating systems.10 Many architectures being supported contrast. The alternative would have been to wait for the international standards to . 1993: p. the international model OSI had competitors that were well established with some 3 5 years’ experience.Telecommunications and networks APPN Application Applications Common Programming Interface and Applications Services Sockets Interface APPN Path Control TCP IP Data Link Transport Presentation Sessions TCP/IP OSI Application Network Data Link Data Link Figure 8. and of integrating their databases with their knowledge bases and making their systems intelligent. Those that had IBM hardware stuck with their SNAs and the upgraded APPN. when 86 network managers had to make decisions on their architectural designs. In contrast. the IS (information systems) managers were in no hurry for the ISDN. Also. 90) APPN TCP / IP ROUTING GATEWAY NETWORK SNA / APPN SNA Hierarchical OSI Figure 8. They had no ideological loyalty to international standards.

The fact that the international marketplace did not accept the OSI model is a source of great sorrow not just from the point of view of network architecture but also from the viewpoint of the development of standards.1: AAPN in HFC Bank HFC is a retail bank in Windsor. says that ‘the cost of adding AAPN to a front-end processor (FEP) is really huge. Sweden SKF is the world’s largest manufacturer of bearings. Robert Amy. has this to say about APPN: ‘It has limited design options for disaster and recovery for us. 1994: pp. 31. .2: Hidden costs of APPN Kathryn Korostoff. principal consultant at‘ Strategic Networks Consulting Inc. . has this to say on the failure and success of OSI: . a further deterrent to the acceptance of OSI was the industry’s lethargy in developing new services and products that were visible to end-users . The competition in the world market gave the consumer and network manager greater choice and pushed the industry into offering a menu of (albeit only three) cost-effective and high performance architectures. ‘More traditional mainframe sites will say ‘‘Why do I need it?’’. . For its 3000 end-users it had an SNA backbone in 1978 with five centres in Europe. Oct. Oct. Source: Datamation. We had to put in extra switches in the event of a failure. Could the OSI model have done just as well if it had a head start like SNA? Do international standards have an inherent sluggishness built into them that prevents them from adapting quickly to changing Case 8. the OSI model being just one of them. John Hogan. 52 3) environments? This is a subject that we shall return to in the chapter on standards. so it would be better if APPN had a way to deal with it’. 1992. it’s that it takes a fundamentally different approach from hierarchical SNA to making applications run across networks. . p. 31. data communications equipment has become in effect multilingual. But a paradigm like APPN will be important for client server applications. SKF has a workforce of about 41 000 working in 150 companies in 40 countries. As a result.Network systems architecture stabilize and be fully implemented. and multiprotocol encapsulation has emerged in today’s router products . . Source: Datamation. p. of Rockland. . SKF went 87 . UK. 1. It is the part of the book that deals with strategies of network management: the set of chapters that is the next subject of our discussion.. . they did not have to wait and adopt an architecture that was universally accepted. because the world moved out from under OSI before OSI could complete its tortuous way through the standards process . you have to increase the hardware capacities memory and processing power’. The acceptance of the layered model served to focus attention on the need to disentangle applications from communications protocols. Korostoff says that upgrading a base-configured 3745 front-end processor to support APPN may cost as much as $26 800 per FEP. the market-place has not accepted OSI . . She says that it will cost an additional $11 800 per FEP in hardware and somewhere around $11 500 in software. Case 8. . who has observed the rise and fall of OSI for over a dozen years. Fortunately for the network managers. 1. It has over 200 AS/400 terminals and is perhaps the largest user of APPN in the world. 1992. (Amy. but without reducing the number of protocol options available to each network user. and that’s the way the world is going. Mass. whilst the network managers did not want to wait in a shroud of uncertainty about which network architecture was to be universally accepted. Usually. the original goals of OSI have been met. This would also have required the network systems and components manufacturers to guarantee the offerings of OSI compatible products.3: Networking at SKF. In 1992. Even the most hardened glass house has got to recognize it’. The manufacturers (at least in the US) showed no inclination for wanting to offer guarantees. . . director of Information Technology at HFC. One lesson that comes out of this rivalry of architectural models is that SNA prospered and survived because IBM managers adapted to the changing environment of peer-to-peer and client server paradigms. Hogan thinks that the basic advantage of APPN is not speed. . Case 8.

In addition. Martin..5 million a year was budgeted a year ‘for the leased lines in the backbone’s meshed core and another Skr 250 000 a year for the routers’. more than 4000 PCs. Redundant links.M. Skr 6. N. (6). B. SNA: IBM’s Networking Solution. But the router backbone cost more that the earlier SDLC (Synchronous Data Link Control) network.. U. used to ensure reliability. Coover. (1988). (1). Bibliography Amy. APPN rises to the enterprise architectural challenge. (2). How IBM is rebuilding SNA. Communications of the ACM. Most of the savings came from staff reductions and ‘elimination of IBM’s hefty license fees for mainframe software’. Tillman.. SNA and OSI: three strategies of interconnection. Data Communications. J. A. D. S. one in Sweden and one in Germany. Shifting SNA onto a global router backbone: SKF shows how it’s done. (1). Deirtsbacher. and Fromme. (1982). 44 52. CNMA: a European initiative for OSI network management. 214 224. Toppling the SNA internetworking language barrier. 9% DECnet and 1% LAN traffic. Prentice-Hall. both using IBM 3090 mainframes.-H. (2) 87 98. The backbone also links Ethernets that support CAD/CAM file transfers. E. Computer Network Architectures. Turin (Italy) and Luton (UK). and Yen. Its backbone is still 90% SNA. Standards by concensus. 58 72. 79 86. R¨ssler. J. S. Routers and more specifically the IP backbone had a distinct advantage ‘because they divert traffic back onto its original path once a line is restored’. and Waeselynck. Systems Network Architecture. also contribute to the steeper charge’. Kippe. IEEE Network. 88 . 33. (Chi-Chung) (1990).Telecommunications and networks from an SNA backbone to a router backbone with 26 routers. ‘That’s partly because it employs higher bandwidth lines to move more traffic and hold down delays. K. (1992). G. and Scoggins. Datamation. Pitman. Gremmeimaier. and leases more than 300 private circuits. M. A. R. Rosen.A. Data Communications. IEEE Computer Society Press. (1987). It closed down the mainframe operations in Clamart (France). Data Communications. Meijer. (20). 28 31. 51 55. Kerr. Systems Network Architecture. over 100 VAX minicomputers and a link to the company’s data centre (King of Prussia) in the USA. Tang. Meijer. SKF uses frame relay to link small sites largely because of its ability to handle bursts. Wiley. R. 37. (1993). 8. Marabini. 22. F. and Peeters. The consolidation saved SKF around 30 million Swedish kronor (approximately £3 million) per year while still maintaining reliability and performance on its corporate network. P. SKF consolidated the five computing centres into two. o (1995). 22. Prentice-Hall. 9. B. Open Networking with OSI. (1992).. (1991). (5). IEEE Network.. (1994). Friedman. A. Source: Peter Heywood (1994). SKF supports over 11 000 dumb terminals.R. 23. (1993).


This Page Intentionally Left Blank .

1. voice and facsimile. This configuration is shown in Figure 9. So what do you do? Well. you have been made an offer that you (and your family) cannot afford to turn down. robust. security officer. But before one does that one must have an organization. Two. Network personnel now reported to a Vice-President of Information Services. Chief Information Officer. In the last part it will discuss all the applications of networking. Peter Stein When you fail to plan. In this part you will learn about management of computing. you have been offered a job as Director of Networking. etc. Here one must identify the different positions needed for network management with the functions that they must perform. They all reported to a Director of IT. This is shown in Figure 9. We shall do this first. You will be responsible for offering network services that are of high quality and performance. However. also known as the CIO.9 ORGANIZATION FOR NETWORKING The illusion of autonomy causes us to ignore the connection with other systems . reliable. The location of network management and of IT depending largely on the organization culture of the organization and also on the importance and complexity of networking to the organization. and accessible to end-users. Robert Schuller Introduction Let us suppose that on the basis of the knowledge you have gained in the previous chapters and with your experience in computing. you are planning to fail. Location and organization of network management In the early days. teleconferencing and text processing. .2. You may make three assumptions. as well as global transfer of data and information for an employee base of 3000 employees spread around the country and abroad. network management was part of IT (information Technology) and was considered as part of support services. much like the librarian. And three. video-conferencing and e-mail. The many functions of managing a network system will be delegated to other staff members and their functions will be identified and discussed in later chapters in the book with forward references to these chapters. . information centre. first you read the rest of this book quickly and carefully. also known as Director of MIS (Management Information Services) or MOT (Manager of Technology). One is that there are no budgetary constraints provided that you can justify the costs. you have a free hand in organizing your department. followed by a discussion of the important function of planning for telecommunications and networking. This organizational grouping recognized the importance of telecommunications and networking to all communications and the desire 91 . The illusion of control causes us to deny the nature of the system in which we are a part. telephone and telegraph. training officers. The one function that you may not wish to delegate (at least at this point) is planning for networking. it became part of communications and was lumped together with other communications related departments like mail. as the need for telecommunications and networking increased and was recognized. The firm has just made mergers and acquisitions of manufacturing facilities both at home and abroad that anticipate a heavy use of networking including applications in multimedia. This can be discussed at two levels: at the corporate level and at the departmental level. standards officers.

Organizations become decentralized and distributed 92 with end-users demanding their own computer access from their desktop to other databases and computer peripherals that they did not have or could not afford.Other OR/MS personnel Technical writer OPERATIONS Operators Schedulers Control clerks Supply clerks Data entry .1 Organizational structure of an IT department Vice-President of Information Services Director of MIS Planning Records Reference Services Library Service Management Research Service Communications Word Processing Typing Filing Retrieval EDP Information Centre Total Systems Planning Information Flow Coordination Data Text Duplication Printing Micrographics Distribution Storage Retrieval Mail Data Voice Facsimile Telephone Telegraph Text Processing Teleconferencing Networking Computer Operation Application Analysis Application Programming Systems Programming Data Administration Figure 9.Maintenance AI personnel . This importance grew as PCs and workstations proliferated and the desire for downsizing increased.Knowledge engineer .Clerks SUPPORT EUC / IC DBA staff Telecommunications staff Security officer Standards officer Documentors Figure 9.Telecommunications and networks DIRECTOR OF IT / CIO / MOT PLANNING Planning staff Policy analyst Problem analyst Technology watcher SYSTEMS DEVELOPMENT Systems analyst Programmer .Supervisors . The end-users became clients of an interrelated computing system served not by central hosts but by servers dispersed not just .2 Horizontal integration under Vice-President for information services to rationalize these related activities and integrate them not only for higher efficiency and lower costs but for greater effectiveness.

The importance of telecommunications and networking can be found in the demand for networking personnel. It shows that only a little less than 40% of all personnel were technical personnel. In addition to an increase in the volume of processing. there has been an increase in its complexity. The high salaries of network personnel is not just a reflection of the high demand for such personnel but also a reflection of low supply.2 Mix of personnel in a regulatory agency Profession Technical experts Information Officers Economists Internal auditors Librarians Lawyers Accountants Technicians Total Source: Telecommunications. Number 8 3 2 2 2 2 1 1 21 are mostly learned on the job. 1991: p. In a firm. The low supply is partly the result of the fact that these personnel have not only to be technical but also to have a very broad based education. Another change in the computing environment is that trade and other daily interactions are no longer confined to the organization but extend across the country and the world. At all levels. At the lower levels. or both? What is the relationship of telephone and mail management to network management? We shall now answer these and other related questions. skills Table 9. This led to the client server systems and downsizing connected by LANs to be discussed in Chapter 10. personnel have to be attracted. Computer processing and its remote communication is no longer confined to data and text but has now expanded to voice and image processing.1 Ranking of the top ten positions in IT by Salaries IS Director Telecommunications Director Data Centre Manager LAN/WAN Network Manager Senior Software Engineer Systems Analyst/Programmer/Project Leader Object-Oriented Developer Client Server Administrator Communications Specialist Programmer Analyst Source: Data Communications. p.1 and is well collaborated by other studies (Chiaramonte and Budwey. Structure of network administration The structure of network administration will be greatly affected by existing organizational culture. A look at the top ten positions in IT in 1994 will reveal that two of the top four and four of the top ten are network personnel. and 93 . A further extension would be their integration into multiprocessing. the ratio of non-technical personnel will be lower but a point that can be made is that telecommunications and networking is a multidiscipline field. Salaries is one indication of the demand for personnel. May 1994. Thus telecommunications and network systems have to be global. 56). This is especially true for senior personnel in a public regulatory organization in telecommunications. Communications are no longer just national or regional but international and global. personalities in top management.Organization for networking in the organization but remotely. which requires large bandwidths and large capacities for wide area networking. They ideally require a national telecommunications infrastructure (to be discussed in Chapter 15) and an international infrastructure (to be discussed in Chapter 16). trained and retained in the highly competitive world of telecommunications personnel. This is shown in Table 9.2. 17 Table 9. The mix of professionals required by Oftel (Office of Telecommunications) in the UK is shown in Table 9. What are these skills? What are the professions in telecommunications and networking and how are they organized in a department? What specialists are required? What is the role of consultants in networking administration? Are they on a retainer or hired ad hoc. This makes the recruitment of technical personnel for networking a little difficult especially at the higher levels of network management where the non-technical and managerial skills are in high demand.

Telecommunications and networks the volume and complexity of the network applications portfolio. It also faces jurisdictional problems that could be very contentious since telecommunications and networking will replace or threaten to replace many existing departments that are large and well entrenched, because they are very labour-intensive. Take the example of internal mail which has to be sorted and delivered. Can that be done by e-mail? E-mail has experienced an exponential growth in the last decade and is the subject of Chapter 19. It may replace much of surface and air-mail, referred to as ‘snail-mail’ because it moves at the relative pace of a snail when compared to the speed of e-mail. This will cause much displacement and some unemployment even though there still will be some traditional mail. Much will depend on the competitiveness of the Post Office and the PT&T as well as the speed and effectiveness with which ISDN is implemented. Another area of contention with networking will be in telephone communications management which is still quite labour-intensive despite much automation of their services. The analogue signals of telephony may soon be replaced partially by ISDN. Yet another area is image processing. This will be partly replaced by ISDN but this is not so labourintensive and will not cause as many contentious jurisdictional problems as with telephones and the mailing room. It is very likely that all these departments will coexist for at least some time in the future but the battle lines are drawn and the battle is soon to come. Whatever the battle, there will be some organizational changes that will be necessary for networking. One configuration is shown in Figure 9.3. Teleprocessing and networking will report to the Director of Information Processing, or Director of IT, or the CIO (Chief Information Officer), who is often a Vice-President. Sometimes an alternative title to a CIO is the MOT, Manager of Technology. This emphasizes the high-tech aspect of the job and is sometimes a preferred title. There are many personnel involved with operations. We will discuss operations in the chapter on network administration, Chapter 13. In this chapter, we will briefly discuss, or at least identify, the other personnel required for telecommunications and network administration. The core personnel would be those involved in development. They will help operational personnel especially on the systems and subsystems that they developed. These personnel would be the network hardware engineer, the network software engineer, and the network engineer, who would most likely be a senior person with a knowledge of both network hardware and software. They will be supported by specialists, like those on voice processing, image processing. In


OPERATIONS Operational Control Maintenance LAN/MAN/WAN Manager Client/Server Manager Installation Manager Voice/Image Manager

DEVELOPMENT Network Engineers Hardware Engineers Software Engineers Specialists Voice Analyst Image Analyst Data Analyst Artificial Intelligence Analyst

SUPPORT Planning Office


Standards Officer

Technology Watcher

Security Officer


Figure 9.3 Organization of telecom and networking


Organization for networking small organizations, the voice and image processing specialist may be rolled into one person specializing in pattern recognition, a subdiscipline of AI, Artificial Intelligence. Alternatively, there may be an AI specialist. In organizations where such specialists cannot be supported for lack of substantial projects, a consultant is engaged for the purpose. The consultant could either be on a retainer or be engaged on an ad hoc basis for consulting on important, complex and expensive decisions. Otherwise the consultants and full-time professionals could be organized on a project basis. Organization by project is not unique for IT and is used for systems development especially in large projects that use the SDLC, Systems Development Life Cycle. Even with prototyping used for development, the project approach is desirable because it involves end-users and management as participants and has a better chance of good problem specification, project implementation and eventually project acceptance. The development staff as well as the operational staff have supporting personnel that include the Planning Officer, which in some cases may well be the Network Manger or Assistant Manager. Then there is the Librarian, a person not unique to IT. So also with the Standards Officer, who unlike duties in, say, Systems Development is not mostly generating standards but enforcing them. Telecommunications and networking is a service that is shared by many others and so standards have to be national or regional if not international. This subject is important enough to deserve a separate chapter for itself, Chapter 11. Security is also a problem known to all of IT, but security in telecommunications is very unique and very important. In these days of LANs, MANs and WANs, the danger of security breaches through communication channels is both serious and complex. It is the subject of Chapter 12. They are related to the operation and management of networks, the subject of Chapter 13. Another support function of any department of networking (and telecommunications) is that of training and education. All these functions have to be planned. The Director/Manager of Networking will participate in this task and in small organizations may be totally responsible for it. This task will be the focus of the remaining part of this chapter. In it we will examine the nature of planning, the dynamics of planning, its process and its implementation.

Planning for networking
Planning is the visualization of the future and taking steps to achieve and strive towards this vision. But is planning for networks any different from corporate planning or even planning in IT? The answer is yes, and no. Yes in terms of the concept and process, but no in terms of the inputs and outputs, dynamics and uncertainty involved. Take for example the IS department. We select this department since it is the highest paid in IT for many corporations and is responsible for all if not most of the mission critical corporate applications. For example, the payroll is critical and affects everyone but this criticality occurs only once a month or perhaps once a week. But for some firms, networking is critical during all working hours as for example the reservation system for airlines or hotels or car rentals. For a firm with world-wide sales, production or suppliers, the working hours may be most of the day, and, for teleworking, it is all hours of the day all year round. And this criticality may be even more extensive as e-mail (and voice mail) and ISDN become more ubiquitous, not just in the corporate world but in all our daily lives. We cannot do much without communications. And so networking is very important and critical. One problem with planning for networking is that it is so dependent on technology which is so unpredictable and unstable. In IS, we have a stable technology and a fairly predictable response from end-users. The telecommunications and networking world is far more volatile and dependent on international standards and governmental actions. The dynamics are different.

Planning variables
A problem in planning is the transformation of long-range goals into strategic and operational objectives. This is especially difficult with a telecommunications and networking (and computing) technology that is often unpredictable. For example, consider the applications environment postulated in the case stated early in this chapter. Does that include image processing? Not explicitly, but if you do know that the firm is in the manufacturing business with factories and suppliers spread around the world, you should predict that the firm will want to send images in the form of drawings and blueprints on the network. Should you plan for image processing? 95

Telecommunications and networks Why not ask management? The answer is that IS has taught us that management does not always know, and if it does know it cannot always articulate its needs in operational terms that a planner can use. There are numerous cases when management has been asked about an application and has responded, ‘No, we will not need that application.’ A little later the same managers demand that application, and if you remind them of their denial of the need of that application, they will quickly reply, ‘Ah, but you should have known my needs. That is why you are paid such a high salary.’ And so it becomes incumbent on the planner to predict the future and do so correctly, or at least close to being correct. examples include the picture telephone that was demonstrated in the 1965 World’s Fair, but the public was not ready for it. Even today, the integration of picture and voice as well as voice along with data is not fully available or accepted in the market-place. A more recent example is with PCs. The industry offered large mainframes but the end-user wanted the PC. When small firms responded, there was demand for them to be integrated. And so increased the demand for LANs. Meanwhile the end-user became more computer literate and more demanding. They wanted decentralization and a distribution of computing power and wanted it to gravitate to the desktop with them being the clients with access to computers acting as servers of databases and other computing services. And so came about the client server paradigm which the industry eventually recognized and then offered the necessary technology for servers and clients. Technology is often driven by the industry and responds to market forces and demands from the industry. Thus the forces of change are both upstream and downstream. But often, and in many countries, technology is driven by the availability of international standards without which there is less incentive to produce products that will have international acceptance. Technology

Dynamics of network planning
The main players in networking are the governmental agencies, standards organizations, the telecommunications and computer industry, and, of course, the corporate world and its end-users. Their interrelationship is shown in Figure 9.4 The important interactions are between the corporate entities (with their end-uses) and the industry (computer and telecommunications). The industry offers a product and the firms and end-users accept or reject it. Historical

STANDARDS Organizations


Telecom/Network TECHNOLOGY

Telecom/Network INDUSTRY Products e.g., client /server system Technology watcher CORPORATE WORLD

Demand for products service Consultants




Figure 9.4 Main players in network planning


Organization for networking is also driven and influenced by governmental agencies depending on the country. In the US, the intervention is not direct but indirect through funding for academia and industry especially in the area of space exploration. In Japan the intervention is more direct with seed-money to industry and the selection of products to be produced and market-share to be achieved. In other countries including Europe there are the nationalized PT&Ts and intergovernmental agencies that have considerable influence. We shall discuss these dynamics in a later chapter, Chapter 15, but for now it is sufficient to say that the interrelationship exists and does influence the external variables in the planning process.
Technological Advances Manager of Networks Compare Plan for & control Operations Compare & control Network Systems Development (Projects)

Planning process
The planning process for networking is conceptually much the same as for IT and corporate planning. Planning is done for varying planning horizons each feeding into the other. Thus the long range plans provide objectives and goals that are to be achieved by strategic medium range plans which in turn dictate the short term plans through instruments like budgets. This process is shown in Figure 9.5. The inputs to this process are many and summarized in Figure 9.6. They include data on external events as much as on the internal environment of goals, objectives and constraints. There is a dependence on advice
End-user/Client pressures

Approval by Corporate Management

Operation of Network

Services Evaluation

Figure 9.5 Implementation of telecom/network plan
Technological Trends End-user needs backlogs future needs satisfaction levels

Political reality: Laws/Regulations People Economic conditions Corporate culture plans power structure Resources Funds Personnel Time

Telecom and Networks Planning Committee

Technology Watcher Consumers Consultants


Figure 9.6 Inputs to network planning


Telecommunications and networks from technology watchers and consultants. The consultants can be a source not only of objectivity but also of a Weltanschauung and global view not often found within an organization. The consultants need to be experts not just on technology but also on the acceptance behaviour of the endusers and customers of networking.
Table 9.3 Possible objectives of network administration Systems applications Should allow for the possibility of the following applications: Distributed processing (e.g. client server system) E-mail Telecommuting Image processing Voice processing Video processing Desirable systems characteristics High speed access to LAN/MAN/WAN Easy access to LAN/MAN/WAN Access should be for asynchronous system (i.e. real-time systems) Masking of complexity for end-user (end-user friendliness of system) Delays and average accesses per successful attempts should be below a specified threshold Low down-time System should be cost-effective Systems should allow for: expansion varying traffic pattern maintainability security survivability

Implementing a plan
There is often a problem in going from a long range set of goals and objectives to a medium term and short term set of operational objectives. This problem occurs because the manager tends to make statements in rather general terms such as ‘very reliable and accessible’ and ‘high quality and performance’. For one thing the variables are not too operational. What is meant by ‘reliable’ and ‘accessible’, and what are ‘quality’ and ‘performance’? The other problem is to specify what is meant by ‘very’ (accessible) and ‘high’ (quality and performance). Lofti Zadeh calls these identifiers fuzzy and has an elaborate procedure for converting these ‘fuzzy’ variables into numerical values (Partridge and Hussain: 1994: pp. 242 53). It is unlikely that the typical manager will have the time to convert networking goals into numerical values, but the planner could state these goals in operational terms. For example, reliability could be stated in terms of delays or response times, like an average delay of no more than 30 seconds or an average response time of 1 minute from request completed to start of response. The values could be in ranges like the average response time of 1 2 minutes. And this may be the start of a dialogue till convergence is reached. This may take a few iterations and one or two plans and planning sessions before the fuzzy values are converted into more operational terms. An example of operational objectives for networking is shown in Table 9.3. It incorporates the long range goals postulated in the Case Study stated earlier. It has added some applications based on what can be expected from the clientele and technology. For example e-mail and voice mail may be desired applications by the clientele but may well be dictated by technology and available whether you want it or not. The performance variables are mostly what is known in IT or have been mentioned in an earlier discussion. One variable survivability may need 98

a definition. It concerns the ability to operate despite a given probability of the presence of disruptive and dysfunctional influences. The concept is related to reliability. If the system is a real-time system, then high reliability is important and this will determine the topology to be used. You may recall that one topology is more reliable than another. And so the objectives of a plan do have an important bearing on the design and operation of the system. Another design decision will relate to accessibility. If, say, high accessibility is required and there is a new building in the design stage, then the long range plan for networking should be to consider making the new building an ‘intelligent’ building with network connections in rooms where professionals would be working. But that is in the long run. How about the medium term and short term? The answer will depend on the distribution of end-users and resources and the computer literacy and attitude of the end-users.

Healy. Good advice. (1994).R. S. 25 28. S. 16(6).R. 3(6).. P.E.-D. K. D. Taming the network: how are telecom managers coping with change? Telecommunications. One of the functions discussed in this chapter was that of planning. 55 59. Data Communications. We also identified many of the functions of network administration and gave forward references to chapters in this book where these functions are discussed in great detail. (1990). IEEE Network. Finding and keeping good people. 257 263. both national and international. (1992). It is such computing paradigm that we will consider in our next chapter. 1995. (1991).1: Headaches for network management A survey of 427 corporate networkers at large global companies were asked to list the ‘risks and threats’ they felt. Information and Management. Torkzadeh. then we may have the potential of a client server computing paradigm. (1989).J. The case for centralized LAN management. McLean. USA). Jan. W. Partridge. Which Computer? 14(6). 58 62. In this chapter we examined the options for the location of the network administration in the organization chart and discussed some of the alternative organizational structures for the network administration department. R. 76 88. Doll. p.M. 8 10. 187 196. D. 21(12). E. Case 9. present. 28(8). The spirit of networking: Past.R. Chiaramonte. Davies.S. We use the term network management for telecommunications management. The percentage responses are 99 . 14 Bibliography Adler. and Peattie. (1991). J. printed in Data Communications. C. software and all the necessary interfaces required. Telecommunications. as follows: Unauthorized systems access Viruses and malicious code Password exposure Internet access Disgruntled employees Information leaks Use of laptops Natural disasters PBX fraud Hackers Terrorism 28% 24% 24% 23% 21% 18% 13% 13% 8% 6% 5% Summary and conclusions This chapter concerns the front-end of network management. Planning has many inputs. They in turn under pressures and influence of governmental agencies as well as standard organizations. which includes not only the network but also the hardware. Desmond. NJ. Flanagan. 31 32. J. 23(6). and Xia. (1992). and Hussain. 341 350. Managing telecommunications by steering committee. We recognize that telecommunications is strongly linked to market forces and the industry of telecommunications and computers. J. bad advice. MIS Quarterly.6. Professional development and the institute of information scientists. 20(4). Journal of Information Science. McGraw-Hill. J. (1994). Smits. Mingay. These are top-down and are in addition to the bottomup pressures from the end-user and corporate management to mask the complexities of the systems and make it end-user friendly without any loss of efficiency or effectiveness. 4(5). KnowledgeBased Information Systems. (1992). P. J. Planning a business communications network. Information and Software Technology. 16(2). G. IT consultants source of expertise or expense.Organization for networking If there is a potential for many end-user clients and many servers of databases and computing resources that are distributed. Managing new MIS professionals. and future. Telecommunications. Source: Datapro Information Services Group (Datran. This is a difficult act for the Network Administrator. and Budwey. 369 379. (1989). and Tanner. and Templeton. (1991). These are shown in Figure 9. 25(2). K.

With this background as motivation. communications facilities. We start with an operational definition of the client server system.10 THE CLIENT SERVER PARADIGM In the 21st century. These servers. however. identify some of its functions and advantages. The number of end-users and clients rose too. mini or mainframe through a bus or a LAN. were computer processors that would facilitate sharing and serve out these resources. requiring not just printer servers. and voice processing servers. Thus came about the servers. Vice-President of Micro Systems Introduction In the early days of computing. there was a backlog of applications and a lot of ‘noise’ between the end-user and the computing providers. In the mid-1980s came the next computing development which was the LAN (Local Area Network). Soon. were 100 accessible to any PC. finally. The client server paradigm that was thus born is the focus of this chapter. Also. But. soon this was not adequate. One was the lack of adequate administrative control and systems management tools. and list its many implementations and success stories. the organizational environment necessary for the client server to be successful. but also image processing servers. There were many limitations to the PC-centric approach. Meanwhile. there were two important technological developments. They must be part of the environment. The number of PCs rose dramatically even within one organization.1. Processors ranged widely in power and price. computers were expensive and computer personnel were scarce. A solution was to distribute the load and responsibilities between the processors at the front-end. we shall examine the components of such a system: the hardware (processor and user interfaces). What was also needed. Such a client server system is shown in Figure 10. and. Development and operations were centralized. the PCs were isolated from other computing resources like expensive peripherals and the corporate database. with the database and application programs residing on them. It is compared with the . and consider the advantages of the client server paradigm as well as the obstacles involved in its implementation. The file server was for data sharing and the printer servers were for sharing printer resources. This enabled PCs (and other computers) to ‘talk’ to each other and share the scarce resources of peripherals and databases. The LAN provided interconnectivity. the human resources need. But. We shall also examine the concern of corporate management with ‘downsizing’ and ‘right-sizing’. (the client) and the back-end (the server). Peripherals became more versatile. Bill Joy. there was a high cost of data swappage and network traffic as well as the inefficient use of the processors both at the end-user end and at the server end. A reaction was towards decentralization into distributed processing where computing was decentralized while much of the centralized computing shifted to the nodes and later to the end-user.2. Components and functions of a client server system An overview of the client server approach is shown in Figure 10. software. it will no longer be sufficient to put computers into environments. One was the arrival of the PCs in the early 1980s with the promise of performance/price ratios higher than the mighty mainframe and yet small enough to stay on the desktop. We conclude with a discussion of the client server paradigm used by many endusers for cooperative processing.

The client server paradigm Data/ knowledgebase Database server LAN Printer Printer server Image processor Image processor server Other peripheral Peripheral server Mini/mainframe End-user PC End-user LAN Figure 10. the access could be through a MAN (Metropolitan Area Network) or a WAN (Wide Area Network). In the case of widely dispersed clients and servers. The corporate database and enterprise-wide application programs typically reside with the server and are accessed by the end-user through the client processor.1 Schema for client server system PC Host computer PC Host computing with master−slave orientation PC PC PC PC Enterprise computer PC PC PC LAN WAN L A N PC Workgroup departmental computer SERVER PC Distributed networking with client−server orientation Figure 10. (The section numbering will correspond to the numbers in Figure 10. each PC is independent for local processing and sharing the centralized enterprise equipment (especially servers) through a LAN. the client server system earlier configuration of a host acting as ‘master’ and the connected PCs (or minis) acting as ‘slaves’. With the client server approach.2 Host computing vs.3. Each of the components of the system will now be discussed in turn. the LAN and the server. as illustrated in Figure 10.3.) 101 .

Typically. Most of the processing is often done by the server and the division of work is determined by a computer program. that is. Thus. We shall have more to say about the user and enduser when we reach the end of the process of using a client server system. a powerful server and a well tuned fast network. we discuss the client. Here is where some confusion can arise. and (through the server) the sharing of computing resources like the corporate image processing system. Whether this means all applications or selected applications or even ‘mission critical’ applications will depend on the confidence in the client server system and the propensity for risk on the part of IT management. Furthermore. one can have quick results with the entire system behind the client being totally invisible and transparent to the end-user who does not need (or care) to know where or how the processing is done. advanced graphic processing systems. not to be confused with the client in computing which is a processor. a colour plotter. it is at least a PC with its own operating system. Sometimes this means rewriting existing applications. which is why 102 programs for a client server system are different from those of a traditional transactional system. when the user or end-user receives the output and results from the system. accessing local files and databases (with or without a DBMS) for browsing and local computing. First. however. or it could be customer. The customer in business and commerce is called a client. an optical character recognition system.3 Components of a client server system THE USER The user is typically the end-user.Telecommunications and networks 1 2 End-user(s) Client Transmission (often LAN) 6 3 4 Server 5 Cable connection Data/knowledge-base and application programs Figure 10. CLIENT The client could be a powerful processor or a dumb terminal with no processing capability. The applications may include message processing like e-mail. a professional or employee of the corporation. The end-user could be a corporate manager. Given a well designed application. who accesses the client for service. . but this client is a human. the client server system is an enabling technology with applications written specifically for it. or just a fast laser printer. these peripherals may belong to a variety of vendors. computing on server-supplied data.

images and animation. By having no programmability. SQL. the spreadsheet. SQL does allow access to multiple database servers providing a wide range of decision-making information and knowledge residing in remote and dispersed locations. The X Windows system is the de facto standard for the larger machines in the UNIX environment. 103 . These de facto standards provide network transparency so that an application run on a server thousands of miles away will appear on the screen of the client as if it were running on the local client processor. It is designed specifically for the type of processing and functions of a client server system with the focus on network communications. images. through which the user communicates.’ (Connor. For a user such as a programmer. At the least. no local diskette drive. the more facilities available. An important component of a client processor is its UI. If it is a general purpose workstation. A solution to the expensive workstation is to have an X terminal. business simulation computing and access to programming languages. a GUI may well be able to accept and deliver information not just in terms of numbers and even text but also as graphics. Smith and making at least $40. it is impossible for anyone to introduce unqualified software at the desktop. However. Meanwhile. thereby becoming a multimedia terminal. et al. greater speeds. As a result.4 English and its Query language equivalent be the word processor. voice. 1993: p. a workstation may be appropriate. are required. In the future. say. e-mail. A GUI. or with more than one X terminal attached to one workstation. However. but for an end-user it had better be friendly or else it may not be used at all. video. including simulation languages like SIMSCRIPT. the de facto standard is Microsoft Windows. most client server systems use a Structured Query Language (SQL) which is a high level dialogue language competing in its end-user friendliness with many a 4GL. which is essentially a workstation without a disk. is the more preferable for the end-user because it enables access through graphical icons rather than through programming commands.The client server paradigm To facilitate the query processing from a client. the UI need not be very user-friendly. but for more complex and large amounts of processing. a workstation can be three or six times as costly as a PC depending on the ‘bells and whistles’ (features) attached. the most popular business programs would English: Find the number of employees working for Mr. even a laptop would suffice. the X terminal’s bitmapped screen allows the application to display all sorts of data formats (text. For a DSS (Decision Support System). and so on) simultaneously. all software on the application server is installed under the quality assurance of the IS organization.4. through a workstation by cable. An X terminal can be connected directly to a LAN. SQL is compared with its equivalent in English in Figure 10. (Socarras. the workstation is faster and has more memory and access to more programming languages depending on what functions have to be performed. AI languages like Prolog and LISP will be necessary. along with its relational database (residing with the server). Graphical User Interface. DBMS processing and business graphics. thereby enhancing the user’s productivity’. though more recent semantics of the access language is increasingly object-oriented. the more powerful the client processor needs to be. the inability to download files and copy them to a diskette. In addition. is almost a de facto standard for client server systems. The X windows protocol ‘can support any number of windows with any type of font or window size. For the PC. drawings. computing power for. the X terminal’s greatest virtue is the lack of functionality. For an engineering workstation one needs not only computing power but also powerful graphics capability for applications like CAD (Computer Aided Design) and systems simulation. 1991: p. Also. For simple data manipulation. 52) It is about time to discuss a network. improves the data security of the network. User Interface. ‘Ironically.000 annually Query for Client processor: SELECT NAME FROM PERSONNEL WHERE MANAGERD`SMITH' AND SALARY<40000 Figure 10. The X terminal is actually an ‘application-specific’ workstation optimized for running the X protocol. graphics performance and an end-user friendly interface. For a KBS (Knowledge-Based System) like an expert system. 52).

Both customers and vendors benefit from interoperability and the resulting competition. or may have an emphasis on data/knowledge-base distinct from being communication-based. the International Standards Organization. The user must pay too by following the procedures and protocols required by the systems. especially those in Europe. despite the availability of software packages. there is often a need for in-house software development. All this included an increased recognition of the importance of the client and the end-user. This is not the vendors’ preference because they would like you to acquire all your equipment and software from them and have a captured market. Yet. but also a user-friendly environment that facilitates and even encourages the sharing of resources. Databases evolved into knowledge-bases and increased not just in number but greatly in types of models and complexity. which in turn is part of a MAN or WAN. not just applications software. who demanded not just fast and easy access. But. is not all that is needed for efficient and effective sharing of computing resources. i. The users must also learn and observe some of the rigidities for using the standard user interfaces and interface languages. This enables a corporation to have an enterprise network that may be strung around the country and yet operate individually. They may be mainframe-centric (centred). The earliest architecture was the SNA but this was proprietary and not open to other vendors. there is the need of fast and reliable telecommunications and interconnectivity. that is the hardware and software can operate interchangeably on each other’s equipment despite their being nonhomogeneous. it is important that there is interoperability. multiple user interface.Telecommunications and networks Network and transmission The server and the client can be connected together by hardwire. However. PC-server-centric. Also. The new paradigm of sharing computing resources has its own demand on specialized supporting resources: it requires a network server OS (Operating System). There is also the need to integrate the client server systems with existing information systems and to use the system not only independently as an end-user but also to work cooperatively among groups of end-users. Since the resources are spread out spatially. in a workgroup. And all this must be implemented while the hardware/software platform and the communications technology being used is transparent to users. objected and supported the OSI proposed by the ISO. One very desirable feature of a client (and a server) is that it has an open architecture. The server is a processor. This was recognized with the increase in the number of PCs when each manager of a PC could not afford the database and peripheral resources needed. Some. they must be connected to a LAN. which is software that enables and controls fast and easy access to databases and application programs. But ISO did not get much acceptance in the US where SNA was accepted by IBM users whilst TCP/IP had many supporters in the UNIX User’s Group. And there was also the need to download some of the applications from the minis and mainframes to the client PCs. though very important. Thus we do not have one universal open system but the systems available are not closed either. a dialogue oriented client server language (such as a version of SQL) and a database architecture. there was often the need to share software. . The importance of resource sharing became obvious. which ensures interoperability.e. This was 104 achieved in the mid-1980s with servers. For this to occur. sometimes a GUI (Graphical User Interface). In conflict with the vendor. SERVERS Connectivity. conceptually very much like the client processor. when they are widely dispersed. it is very different. not just in one country but also abroad. or at a department or local level. Much of the software required is available from vendors and software houses and the client server systems vary greatly in emphasis and capabilities. such as the one for SIMSCRIPT 11 for the occasional user of simulation. the operation and exchange of information in a heterogeneous mix of equipment and software using the network. For one thing it does not have a UI. The essence of openness is that there is interchangeably of components of the system and therefore vendors have to compete with better products and better service. Interoperability also implies an open network architecture. it is in the interest of the customer to have the flexibility to mix and match components of the client server system. but also compilers for programming languages.

In any case. Application tools and lower prices have made client server systems development more competitive in terms of cost performance when compared with mainframe and minicomputer development. ž Bulletin Board access. This should be done without affecting the access for legitimate sharing and without an uncontrolled proliferation of islands of data cropping up everywhere. ž File sharing. Which type of server is desirable depends on the needs and objectives of the system. The platforms for sever processors are PCLAN servers with mainframes and minicomputers as alternatives. Most of the processing. But the client server application development does differ from traditional software development in some significant ways that include: 1. Access optimization. it may perform data retrieval. Because of these capabilities. ž Batch processing. but should not impact on the consistency of data and the responsiveness for the end-user. The enterprise data that is a corporate resource . The front-end client portion is run by end-users using languages like SQL that have simplified data request protocols and extract data from whereever it might be located. Applications processing Data is used by application programs and most of these reside with the server. small businesses or professionals like doctors) who cannot afford their own processing to use a client processor. Still. and which ever operating system controls it. There are some off-the-shelf client server applications now available and these are increasing in number and scope. or. access to data must be controlled to maintain security. say. in which case it is a database server. 105 Database processing By processing data at a server instead of at a mainframe there are some principles of processing data that are applicable. These functions include: ž Network management. Why are servers so expensive? Because they perform many functions. servers are much more expensive than a client processor. disk space utilization. ž Storage. The server may differ as to their responsibilities and functions. security utilities and I/O (Input/Output) handling would vary with servers manufactured by different vendors. in the range of $20 000 to $50 000 at 1995 prices.The client server paradigm let alone a GUI. retrieval and management of documents. in which case it is a file server. ž Facsimile transmission. ž Gateway functions. thus allowing the processing of customized ‘wild’ queries. including access to outside access and public e-mail. needs to be unified and integrated. Much of data management (and resource sharing) is automated. For example. This enables multiple users (such as. and fault tolerance. database processing and applications processing. whatever computer stores it. This improved responsiveness is important. backup and recovery. Another important issue is the management of data (and knowledge if any). however. use multiple operating systems. gateway and other access control. the server may act as a repository and storage of information. many applications must be painfully developed from scratch. Some of this is done through a DBMS that resides with the server which controls access between multiple processing systems (and even multiple distributed databases) and integrates data access with network management. It is different from a general purpose processor and is sold as such: a server processor. These include the issues of integrity. the server must be able to do multitasking (perform multiple functions simultaneously). is data entry and could still be done in batch with updating done in the day and processing done at night and transmitted by LAN for use in the early morning at the local client end. and recovery must be possible after systems failures. The server platform and issues relating to the servers are hidden from the end-user. It is designed for networking. data management. security and recovery of data. especially for a query which could now be a fraction of a second instead of 3 5 seconds under a mainframe. have scalability (the ability to upgrade upwards without loss of software performance). however. Processing functions are distributed between client and server. but there are issues which include: internetworking. and have a fast response time despite the time required for teleprocessing. be portable.

5. are used. but the net is a decrease of overhead costs of computing at the centre. the programmer is concerned about the loss of integrity and security of data. Why? Because the programmer is concerned (justifiably or not) about the loss of control over development. 37). 6. While the professional computing personnel (including IT management) are not enthusiastically in favour of the client server system. But. if not most important to the end-user. The end-user gains greater control over the system. Also. Some cost components increase and some decrease. UI and more often GUI are used because end-user friendliness is still very important. but not necessarily a reduction of total costs. The manager is concerned about a loss of power through a reduction of span of control and a loss of control over the acquisition of resources. downsizing in management parlance is the reduction of the labour force and the firing or displacement of personnel. Development tools like SQL Windows. processing and use of information. Progress. unlike the end-user. and Uniface for OS/2 are emerging. 1994: p. All this is done by the programs residing at the client. It is the decreasing of their overhead costs and is downsizing for them. Support for the client server system comes from the end-user who can see the reduction of time and ‘noise’ and an increase in responsiveness in the development of systems. The annual costs per user in the fifth year (with 3584 mainframe users) in one study came to $1484 for the mainframe and $2107 for the client server system (Semich. End-users are now more computer literate and experienced and no longer fear the computer which is becoming more robust and friendly. The right size and right timing is referred to as right-sizing. like the DBA (Database Administrator). SERVER BACK TO CLIENT Once the applications are processed and data retrieved as per request by the client. in terms of computing resources there need not be any change in labour force. However. There are other organizational consequences of the client server systems: downsizing. Thus far we have examined simplified line diagrams of the client server approach. This is then given to the user through the UI. This shift is welcome to corporate management at the centre though they do share the concerns of losing control over the development and processing of information. favour. Advanced networking. There the application results may have to be formatted for better display or the data retrieved has to be processed. the costs for the client server system is higher than the mainframe in total costs though the initial costs are lower. though OO methodology is being increasingly used. 4GLs and code generators are used extensively. User Interface. This is popular in recessions and bad economic times while upsizing and expansion is popular in good economic times. mostly LANs. But. The downloading of computer processing from the mainframes (and minis) is referred to as downsizing by corporate management. The programmers and the DBAs are supported in their concerns by the Manager of IT. 4. This coalition of corporate management and end-user with help from the computer industry (in the delivery of appropriate systems) will overcome any resistance to the client server system which will be increasingly implemented throughout business and industry.5. A formal schematic diagram of a real-world application is shown in Figure 10. The client server paradigm is largely minicomputer. There is a reduction of overhead computing costs at the centre. They are willing to take the responsibilities of controlling the clientend and leave the server-end to the centralized processing centre. the corporate management is enthusiastically in 106 . CASE tools are being used with Rapid Prototyping. the results are sent back the way they came. ObjectView. FLOWMARK. through the LAN and the client processor.or microcomputer-centric and is competing well with the mainframe-centric systems where most of the ‘mission critical’ applications reside. 3.Telecommunications and networks 2. powers and responsibilities that now gravitate to the department administrating the client node. Organizational impact A user who may be a computer programmer. may well be less than enthusiastic with the client server paradigm. but for an additional reason. What we have instead is a shift of costs (and responsibilities) to the client nodes. Why? Because there is an opportunity of reducing costs at the centre.

The client server paradigm Figure 10.5 A real-world application of the client server system .

the server and the end-user. then you will be somewhat happy at your reduced responsibilities for the system but you are now concerned about the integrity and security of the system. in implementing a pilot study. Some have been implied in the discussion of the organizational implications: problems associated with downsizing and the need for greater coordination and cooperation among the end-users of the system. End-users have to be trained not only on using the client and knowing the functions of the server. the shortage of experience in the planning and implementation of a client server system. The less risky approach is to implement vertically and do so on a pilot study basis. These advantages are summarized in Table 10. So. the server. . It will be sometime before IT personnel and corporate management are fully comfortable giving up their centralized processing. This approach is also flawed in that some resources lie idle while others are being developed. The other approach is to implement horizontally. Other technological obstacles are the lack of tools of development and products of the client server system. The third alternative is to do all at once. especially with increasing open architectures in hardware. the clients. On the whole. especially the need to train personnel in the use of the new system. It is likely that the third approach is often not feasible because of the high investment necessary in money and skilled personnel. and an experienced IT staff running the development and operations of their information and knowledge base. They want more control over their operations and the time required for development. There is also the problem of resistance. a LAN for one business does not make economic sense because once implemented a LAN can be used by other applications.e. implementing. Bad economic times and recession are allies for the client server system. for example. But endusers are taking the initiative and the responsibility. open systems in telecommunications. there are some obstacles and risks. Training may be considered a technological obstacle. It is also highly risky. the lack of methodologies. risks and limitations involved. horizontally and vertically. Obstacles for a client server system In achieving the advantages.1. and the lack of national and international standards relating to the equipment and operations of the system. One is to go vertically and implement all the levels of implementation. but end-users need to be educated about networking and trained in navigating across the LAN and perhaps even the Internet. it is desirable to select an application that is less risky and has low costs and low visibility. which must be expected of all new technological changes. its back-up and recovery practices. there is a learning curve involved. especially that of the data/knowledge-base. There are also problems of conversion. If you are a corporate manager you will be thrilled with the advantages of downsizing and the reduction of costs at the centre. If you are an end-user you will be pleased with getting control of the system even though it may mean that you cannot blame computer personnel for things going wrong and instead have to take certain responsibilities for the system (at the client-end). Management wanting to downsize and reduce their centralized overhead will be more willing to give up some of their centralized control to the client. The first implementation in one case cost $2 million. The technology will work in favour of the end-users and the corporation and make the computer industry more competitive and responsive to the enduser’s needs. but for only one business at a time. its fault tolerant mechanism. Also.Telecommunications and networks There are three alternatives in implementing a client server system. Downloading to the end-user’s desktop is perhaps inevitable. 108 Advantages of the client server system The advantages of a client server system are context-sensitive and depends on your vantage point. especially at the management level. applications and the end-users (their training). but then the next implementation of similar scope was almost half the cost. However. viewing the system from the point of the enterprise. If you are one of the computer personnel. i. the LAN. and the refinement of multimedia equipment. there are more advantages overriding the disadvantages. which in many cases is fairly well stabilized with its security regime.

6. . to database servers. spreadsheets and e-mail. Some of these productivity tools. the server facility. We conclude with a view of future paradigms of client server systems moving from the Ethernet era (of file servers. analysis.The client server paradigm Table 10. geographies and time zones Rewriting systems for the client server system is often an opportunity to purge obsolete software from the application portfolio and to consolidate and integrate the system. 1995: 109 . one or more clients. The end-user can do some local processing on a local database at the client facility and also use the many software productivity tools such as word processing. . groupware and transactional processing monitors) to the intergalactic era of the late 1990s and early 21st century with object-orientation in distributed processing (Orlafi et al. where we see that an end-user navigates through the client facility. The client server system strips data off transactional systems and stores it in the server to be shared for analysis and even manipulation locally Enables distribution of processing from centralized to desktop computing Offers cooperative processing between individuals and group departments across organizational boundaries.1 Advantages of a client server system Table 10. and presentation.2 and can be overcome only by close cooperation between IT personnel. and make it more efficient Offers more friendly interfaces for end-users. corporate management and end-users. cooperative processing does increase the cross-currents between end-users and thus raises new problems of collaboration and cooperation in intercommunications and may lead to adhocracy. the reverse path is taken back to the end-user. like e-mail. as well as different times and different places. The plug-and-play possibility applies at least in theory when parts of a system can be replaced without impact on the rest of the system These obstacles and others have been summarized in Table 10. The resources and infrastructure needed for the client server paradigm is displayed in Figure 10. Productivity tools extend communication to different times but the same place. along with the underlying operating system and interprocess communication systems. especially knowledge workers and customers Greater involvement of end-users in IT implementations The open architecture and open systems offer flexibility in choosing different configurations of hardware. network and DBMS from multiple vendors There are greater possibilities for expansion by adding hardware (even laptop computers) to networks without replacing existing hardware.2 Obstacles in the way of a client server system Organizational Lack of personnel skilled in the client server system and in networking Resistance to change and new technology Risks of downsizing Costs of conversion Need for greater coordination and control of more end-users Technological Need for LAN/WAN infrastructure Lack of skills and equipment resources Lack of methodology and experience in planning for a client server system Lack of client server products and tools of development Lack of client server applications Lack of national and international standards for the client server paradigm.. the network infrastructure. However. Having done the desired computing. and one or more servers. Summary and conclusions Alok Sinha has described the client server system well: . Reduction of responsibilities and cost overhead Better local cost control of operations and development (original and modifications) Faster response time to requests for processing Greater access to corporate data/knowledge otherwise maintained in a highly protected and centralized data structure. will be used more intensively in cooperative processing which connects the clients horizontally and improves horizontal communications. and then to the applications and data/knowledge-base. form a composite system allowing distributed computation.

good contingency and a platform that would be scalable to the required processing power.3 Paradigms of client server systems 1982 1986 1986 1995 FIRST WAVE (Ethernet Era) File Servers SECOND WAVE (Ethernet Era) Database Servers Groupware Transactional Processing Monitors THIRD WAVE (Intergalactic Era) Object-Orientation in Distributed Processing 1995 20xx There was token ring LAN connecting PS/2 (PC computers) at 16 sites with clients and servers at over 3000 sites. 122). p. The objectives of the system were to have a very high level of fault tolerance. This evolution of the client server system is summarized in Table 10.1: Client server at the 1994 Winter Olympics IBM installed a client server system for the Winter Olympics in 1994. DATA/KNOWLEDGE Figure 10.Telecommunications and networks End-User CLIENT Network SERVER APPLICATION PROGRAMS. 2000 athletes. They deliver transactions to other Case 10. It was designed to serve 100 000 potential users that included 50 000 accredited personnel. There was also an IBM ES/9000 mainframe which served as a central database with over 250 000 files available for the network. The OS/2 graphical user interface (GUI) offered easy access to all users accessing a client.6 Navigation in a client server system Table 10. 8000 media representatives and over 100 000 visitors per day. Citibank runs Windows NT on Compaq servers. there was the RISC System 6000 that was used for the design and planning of the games.2: Citibank’s overseas operations in Europe Citibank started moving workload from its 176 mainframes in 17 countries to a client server system. The application was ‘mission-critical’ and reliability had to be absolute since there would be no second chance to capture say a ski slalom race or a lug run. The client server platform handled 30 000 transactions per day with a value of approximately $200 million. In addition. And what if the results were inaccurate? That would not only cause a flurry but bring many an athlete and relatives to tears! 110 .3. The network architecture used was IBM’s SNA whilst the PS/2 handled all the accreditation and games management. Case 10. The strategy was one of re-engineering to have a more open and flexible platform that would support an object-oriented network which would provide information for enhanced decision-making and lead to higher productivity.

reports its right-sizing effort to result in halving its costs under a mainframe (Connor. Datamation. P. and von Schilling. W. Heinz Pet Products used the client server paradigm to increase productivity and improve customer service.. The right way to rightsize. Information Systems Management. improving customer service and bringing the product faster to the market. The bank worked with Digital Equipment using a integrated suite of software products for enterprise information delivery developed by the SAS Institute. Citibank’s big gamble. increasing productivity.A. The Race for client/server CASE. J.H. M.B. (1994). Computer 27 (5). 1993: p. 28 (3). 1993. 11 (2). 35 (7). GE in one application showed a one-year payback for its start-up cost through its downsizing effort using a $6000 PC. 277 302.5 billion manufacturing enterprise. PC Magazine.) (1994). T. a large manufacturer of computer systems. Unisys. Lessons from three implementations: knocking down the barriers to client/server. do bookkeeping. P. A. Watterson. H. R. (1993). Plans and policies for client/server technology.-P. (1994). Information Systems Management. (1991). a third and improved service to end-users by implementing the client server technology. Benefits and barriers to client/server computing. and check writing. J. D.S. Corporate downsizing and new technology. D. and Chen.J. Canning. Harkey. R. N. enhancing decision-making support. J. I/S Analyzer.-C. Intergalactic client/server computing. reporting. N.E. (1992). IEEE Spectrum. 93 96. (1995). Hsiangchu. 20 (4) 108 122. DSS support.The client server paradigm banks.D. K. Pinella. By 1995. with assets of $116 billion. Human Resource Management. 181 298. Client server meets the Web generation.S. May 45 55.F. (1992). accounting. pp. Electronics Distributor built its client server system incrementally for tracking marketing information. OOP. N. Seymore. 7 24. J. It used the Hewlett-Packard HP 9000 Business Server. and Stonecypher. (1994). Client server computing. Communications of the ACM. 32 38. Byte. (1993). (1997). Anatomy of an X terminal. The Bank of Montreal in Canada. 13 (3). Liang. 15 22. (1994). Semich. Special Issues on Downsizing.A. 51 54. 12 15. 30 (4) 1 16. K. 37 (20) (1993). delivering transactions in an electronically structured format. Application development tools: client/server. When client/server isn’t enough. and Edwards. Bibliography Cameron. Cover story on ‘Client/Server Computing’. Schultheis. 38 (5). 11 (2).. and CASE. 52 55. (1994). 10 (2).3: Applications of client server systems The Morning Star group used the client server approach to downsize from a 15 year old mainframe system and achieved a faster time-tomarket response and greater productivity. Which Computer? May. drafts. Miranda. Journal of Systems Mangement. Cooper. R. 36 43. Muller. 73 79. 33 (2). plans are for all locations in Europe to be linked by a wide area network and document imaging system tied directly into the new platform. 23 27.-S. Orfali. (1992). Datamation. and Tellerman. 2(2). Application server at your service. 77 98. 45 (2). Chen.. A.. Socarras. Information Systems Management. (ed. Source: George Black (1994). M. MIS.-S. and Bock.W. cut its information system’s costs by 111 . re-engineered its operations to provide innovative and timely solutions. Wei. Sinha. a $3. (1994). UNIX Review. 10 (16). Levis. W. Motorola. 42 44. Case 10. Connor. Can you orchestrate client/server computing? Datamation. 56).

but which is to be improved tomorrow you get somewhere. equipment will not be interoperable. Telecommunications are also needed in our world today because we are in an age of multivendors and multicarriers where interconnectivity and interoperability is . 1994: p. First. but it did not get public acceptance. Sometimes. So what is so tortuous about international standards? Are national or local standards any less tortuous? What are standards anyway? Are they necessary for telecommunications and networking? If so. Some standards are crucial like when lives could be lost because we do not drive on the same side of the road in all countries of the world. in the Japanese case we look at the national organization and its decision-making process. that of the B-ISDN. 52). These are the Europeans. In the European case we examine a regional organization and its standards-making process. These are standards we all unconsciously acknowledge. largely because it did not ‘complete its tortuous way through the standards process’ (Amy. But these are not very crucial. we look at standards. Sometimes things in life are not as standard as we would want. International Standards Organization. like when we travel with electrical appliances and our plug does not fit the socket in the wall or we do not have the right frequency of electrical current. This is very true in telecommunications where. 60 minutes to an hour. Henry Ford Introduction We discussed standards when we described the OSI model. and in the American case we look at a case study in which an international standard was developed. But as the world gets interrelated we need standards that are not only national but international. though. It was to be the international standard of network architecture. 112 What are standards? When measuring quantity. their need and their rationale. we will look at a different perspective to give us a balanced and ‘total’ picture of the process. the Japanese and the Americans. however. or even the number of days in a year. like the day of rest in the week. transmission media may vary and international communications will be impossible. Why can we not have one shape of plug that fits in all sockets and have just one frequency all around the world? This was not necessary in early times when people did not travel with electrical appliances that were crucial for shaving or drying hair. Standards are accepted authorities or established measures of behaviour. without international standards. In each case. protocols will be different. we adhere to standards such as 12 to a dozen. a lack of standards is just a big pain in the neck.11 STANDARDS If you think of ‘standardization’ as the best that you know today. time or distance. We will also look at the national and regional organizations that have done most for the development of international standards. what are their advantages and limitations? What does it cost? What is the process of standardization? Who formulates them and who enforces them? It is these and related questions that we will examine in this chapter. we look at the development of standards for the OSI model at the ISO. For a view of decision-making at the international organization itself. operations or performance. This will affect the community at large but definitely the business community doing international trade.

chips and wafers. encourages competitiveness. stimulates demand. telecommunications has become more ubiquitous and standards have been developed in areas including specifications for wiring. We need international standards if we are to utilize the potential of telecommunications and increase world trade and international interactions. Instead there is a National Institute of Standards that must react to the needs of standards. modems. operating systems. connectivity devices. However. Theoretically it is all the countries in the United Nations that must agree to an international standard. Without such standards we would not have the interoperability we now have.1 Topics under consideration by subcommittees of the International Standards Organization. and slow technological progress because producers will create duplicate products instead of building on previous work. In many countries there is a nationalized communications PT&T and a government that can dictate national standards. systems development. These standards are part of the work done by international standards organizations like the ISO. spurs innovation. getting agreement among these developed and interested countries is still slow and difficult ISO/TC97 Computers and Information Processing SC7 Problem definition and analysis SC1 Vocabulary SC2 Character sets and coding SC3 Character recognition SC5 Programming languages SC6 Digital data transmission SC8 Numerical control of machines SC9 Programming language for numeric control of machines SC10 Magnetic disk packs SC11 Flexible magnetic media for digital data interchange SC12 Instrumentation tape SC13 Interconnection of equipment SC14 Representation of data elements SC15 Labelling and file structure SC16 Open end connection SC17 Identification SC18 Text and facsimile communication SC19 Office equipment Figure 11. but in practice it is a few industrialized countries that do the work and have the expertise. Lack of international standards will hamper the marketing of products beyond their national borders. The existence of standards not only improves communication. Technical Committee 97 113 . The rest follow. International Standards Organization.1. but also provides portability. In the early 1980s. We look later at just one of these. there is no ministry that can dictate standards. and so on. and provides an organized way of sharing and transferring technology. Since then. helps planning and control of operations. Standards in telecommunications is part of the problem of standards for the computing industry where standards are needed for programming languages. interface devices. data representation the list can go on and on.Standards necessary and this can be achieved only through international standards. reduces uncertainty. OSI network architecture and protocols. Note that there was some interest and acknowledgment of interconnections and communications. the OSI. reduce information exchange between customers. database and knowledge-base design. But international standards are far more difficult to achieve than national standards. ISDN. it is easier to agree on standards than to have international standards because the players involved are global and greater in number and diverse in interests. But even in the US. In countries like the US. manufacturers and retailers. transmission. they were active in many areas like those shown in Figure 11. voice and video processing.

trade associations and professional organizations. it is now a global market-place also being used by the household and individual who uses telecommunications for e-mail and entertainment. upgrading and multi-vintage compatibility.4. the functional standards are then tested.3. Not all parties and all players are involved all the times. In some countries the carriers and manufacturers are the same. the two have split and now compete with different vested interests. promotion groups and user groups. These base standards develop into functional standards also called profiles. We can see this in Figure 11. Testing is also concerned with compatibility and at different levels: multi-vendor compatibility. They represent the market forces. The process of decision-making for standards is both upstream. The standardization players are shown in Figure 11. and downstream. The organizations that reigned supreme in the early years are now giving way to the technological forces which in turn are giving way to the market forces that are largely the consumer. consumer groups. Hence international standards have to be increasingly conscious and sensitive to these needs which 114 makes the mix of the market more heterogeneous and difficult to predict and to cover every aspect of suture intended applications.2 Players in Developing International Standards because they may have conflicting objectives and interests.2) is changing as shown in Figure 11. The relative strength and influence of players in telecommunication standards (shown in Figure 11. After feedback and participation by regional and national standards bodies.2. Different players and parties are involved in different stages in the development of standards. From what was businesses using telecommunications for national trade and commerce. and contain a limited subset of the permissible variants. Testing could be done by independent organizations where the emphasis is on conformity with the design specifications of the standards. where the base standards. Another set of players includes the consumer. This was true in the US till recently when Western Electric produced equipment for its parent company that was a carrier: AT&T. are proposed by an international organization (with participation by other players at a high policy level). and product-line compatibility. In parallel. This consumer driven sector is also changing. the technical personnel from countries that are members of the international organization also perform testing but with emphasis on interoperability and performance first and conformance second. from the corporation and user to the international organization.Telecommunications and networks Organizations • nations/countries • international • regional Equipment manufacturers and suppliers Network service providers and carriers Others • promotional bodies • user groups • professional groups • market (consumers) Figure 11. With deregulation laws. from the top down to the user and . This includes not only carriers but also manufacturers of equipment that are directly part of telecommunications or use telecommunications like the computer manufacturers which are far from being a homogeneous group. One set of players is the computing and telecommunications industry. containing variant and alternative methods that are implement-defined facilities.

the players involved are 115 . both upwards and downwards.2). standards proposed by ISO are accepted by the globe. Fortunately.4 Decision-making process by international organizations corporation.Standards Driving force Market driven • consumers Technology driven • carriers • suppliers • computer industry Organization driven • market • user groups • Professional groups • promotions groups 1970 1980 1990 2000 Figure 11. in the development of standards for a modem or say a cable connector. At other times. not even in the negotiating process where only a few sets of players are involved. there is much feedback all along the way. not all players have to agree on everything all the time. an international standard is not adopted by all countries as in the case of the OSI model. For example.3 Forces Driving Standards Professional organizations Trade organizations User groups SIOs (Scientific & industrial organizations) Technical Personnel from member countries Base standards Inter. Sometimes. It is not always possible or desirable to have complete agreement by all parties (the innermost and shaded part of Figure 11. organizations Regional Stand bodies National Stand bodies Functional standards RPOAs (Regional Private Operating Agencies) Testing • performance • conformance Testing • conformance • performance Independent organizations Analysis Acceptance by international bodies Yes OK ? No Figure 11. Ideally.

the OSI has been adopted in Europe but not in the US. the OSI model. Starting from scratch was what happened with ISDN and it is one of the many success stories for ISO. the ISO came on to the scene too late and took too long. Sweden. An example would be the standards for network architecture and protocols. a very popular operating system. Canada. This delay is partly due to the fact that the OSI was so comprehensive and important. governmental bodies. Again. UK and US) Other (industrial representations) OOS POSI SPAG X/Open On conformance On interoperability ANSI (US) BSI (UK) DIN (Germany) TTC (Japan) I S O PROMOTIONAL BODIES TESTING GROUPS NATIONAL BODIES Figure 11. The testing was done in parallel. was also popular and was based on UNIX. Canada.5. One is that OSI had to compete with models that were operational in the US. In the case of the OSI. Sweden. Most of the government participation came from Australia. By the time that the standards were ready. One of them the SNA was developed by IBM and it was difficult to persuade IBM that their design was not the best. As mentioned earlier. The standards were developed in workshops with feedback from numerous groups and tested by yet another set of personnel. The profile bodies had representation from regional bodies around the world as well as from international and national PROFILE GROUPS National Institute of Standards and Technology OSI Implementors Workshop European Workshop for OSI Asia Oceanic Workshop Governmental Groups (mostly Australia. Also. the environment had changed: PCs had proliferated. It is perhaps appropriate to discuss the process of producing this standard for it is a technology that we have discussed in previous chapter. The assumptions underlying the start of the standards specifications The development of OSI It took ten years to develop standards for all the seven layers of the OSI.5 Players for work on OSI 116 . There are many reasons for this.Telecommunications and networks the suppliers and carriers and the international standards organization represents the remaining parties. But there are some situations where many players are involved and in a very serious way with high stakes.5 are somewhat different from the generic names used in earlier figures but are not inconsistent. Another set of protocols. distributed processing had become popular and LANs had emerged. UK and the USA. Timing is highly critical for the success of international standards. The organizational structure and personnel involved are shown in Figure 11. The names chosen for Figure 11. They developed the base standards which were then passed on for feedback from numerous promotional bodies. Many of their concepts were adopted by OSI but it is more difficult to build on an old structure than it is to start afresh from the bottom. one for conformity and one for interoperability. and partly due to the thoroughness of the process of developing an international standard. the standard can have a far-reaching and important impact on global telecommunications as well as being an interesting and controversial case from the organizational point of view. TCP/IP. The second reason perhaps is that it took too long for the international standard to come out in what was a highly volatile technological environment. this is not conceptually inconsistent with Figure 11.4.

In 1993. CCITT is part of ITU (International Union Telegraphique) that was founded in 1934 and in 1947 it became an agency of the UN. directors.Standards had mostly changed during the long period of gestation. each exercising one vote. 14) International Regional ETS (Europe) TI (North America) National Special industry groups Forums ANSI Private Industry Local Leadership role Follower role Figure 11. One indication of the work of the ISO and the growth of ISO standards can be seen from Figure 11. advisory groups. voice network operations and maintenance. It has 15 study groups including those on transmission equipment and service systems. International organizations have a difficult task of determining when to start on standards. 14. Not all these are related to telecommunications but many are.6. the ITU had 165 members. The ISO (and its successor organization) have a vertical hierarchy of decision-making involving its international membership at the regional and national levels. Oct. some directly and some indirectly. It shows the maintenance of 40 international standards in 1992. 1993. This long period was partly due to the fact that it was an international standard prepared by international organization and this process just takes a long time. Working parallel to the ISO is the CCITT (Comite Consultatif Internationale Telephonique et Telegraphic). 1994). modems. and open systems. switching.7. The national levels have their own hierarchy. Why? To answer this we must look at the organizational structure and process of international standards organizations. Mostly bodies like governments that were concerned with telecommunications or state owned PT&Ts were members. In 1991. but it is only one of many ISO products. ISO/CCITT The ISO The OSI model may not have been too successful (in the US. study groups.7 Heirarchy of effort at ISO 117 . One configuration of the hierarchy of effort in the US is shown in Figure 11. that is). 1989 1990 1991 1992 Drafts for Publication Drafts for voting Published standards and technical reports Standards being maintained 33 0 0 0 116 21 18 0 90 104 47 0 106 103 184 40 Figure 11. The UK is represented by the Ministry of Trade and Industry and the US by a mix of representatives from both government agencies and suppliers but coordinated by the Department of State (Foreign Ministry). as well as regional conferences and international conference (Irma. languages for telecommunications.6 Growth of standards and technical reports (source: International Herald Tribune. The international body has its own internal hierarchy which includes councils. p. a structural reform of the ITU saw the demise of CCITT and its resurrection as the standardization sector of ITU (ITU-T).

These are summarized in Figure 11. 118 European standards organizations The European standards organizations are shown in Figure 11. Radio Consultative Comm. Not identified are the many interrelationships with the CCITT.9 with the important relationship down under being identified. We make the choice largely on geographic distribution. JTCI = Joint Technical Comm. ISO and the CCITT have numerous organizations with which it has a lateral and horizontal relationship. You may still be in the dark as to the type of interactions that take place at the national and regional level. Figure 11. The diagram also emphasizes that each one of them has a potential input through national organizations.Telecommunications and networks National Regional Global ISO/IEC ITU T T C C C I R CCITT JTC1 IEC IEEE DIN ANSI BSI CCIR = Inter.8 Organizations involved in International Standards You cannot develop too early before the technology has stabilized or the standards will not be relevant because they are obsolete. these are the organizations and countries that are most active in the formulation of standards.9. though some national bodies like the national standards bodies and national telecommunications administration do have additional direct links to international bodies as indicated in Figure 11. the connections are through trade associations and user groups at the same level as our government. Electro. You cannot start too late because then vested interests dig-in with a loyal following and universal acceptance becomes difficult. It has 269 members with much of the detailed work being done in project teams appointed for a defined task and for a limited period of time. Other inputs to the European Standards Bodies are from National Standards Bodies and the National Telecommunication Administration. we discuss one regional (in Europe) and one national (in Japan) organization and see its structure and decision-making process. The ETSI (European Telecommunications Standards Institute) is the body that issues the standards. 1 IEC = Inter. Technical Comm. Coincidentally.8 to give a picture of the many organizations involved which may explain the sluggishness of international organizations. The choice is yours. The proposals of the project teams . To throw some light on these processes. the ISO and other international organizations. If you wish to influence an international contract you will have to contribute your time and effort. Thus both the private sector and the public sector as well are represented. At the national and corporate level.

DRAFT Standard Experts Analysis Nationl Bodies Public Consultation Members (weighted voting) ASSEMBLY European Standards Institute Figure 11. while still providing a measure of stability so important for implementors.Standards ISO ITU CEN CENELEC ETSI CEPT National Standards Bodies National Telecommunications Administration Manufacturers Public Network Operation User/Service Providers National Administration 63% 14% 13% 10% KEY: CEN = European Committee for Standardization CENELAC = European Comm. ‘In this way. concepts can be put together quickly and adjusted to take account of changes. 61). for Electrochemical Standardization ETSI = European Telecommunications Standards Institute CEPT = Conference European des Administrations des Poste et des Telecommunications Figure 11. Proposals are base on the ‘best’ estimate of technology.10 Standards-making process in Europe 119 .9 Regional standards body in Europe goes to the technical committees and is then passed on to the technical assembly for approval where a weighted voting method is used.10. The ETSI uses the ‘working assumption’ process as shown in Figure 11. 1992: p.’ (Mazda.

The standards not only have to be feasible but they must also be acceptable to the computing industry and the important governments involved. Little of this cost is borne by the sale of standards and most of the cost is borne by a few countries. It is quite different from the other side of the Pacific. Once the right standards are in place at the right time. Summary and conclusions International standards in telecommunications is at the confluence of two sets of dynamic 120 forces: the telecommunications industry and the telecommunications environment. Their process of decision-making is shown in Figure 11. EDI (Electronic Data Interchange). in the US. evolving switching technology. the demand for distributed processing and the client server system. the standards must be in place at the proper time: not too early lest an undeveloped technology gets ‘frozen’ by the standard. Council and Technical Assembly to which one Research and Planning Committee and five technical subcommittees (TCS) report. the persuasiveness of wireless technology. TTC is also a member of the regional INTUG which includes memberships from Australia. Standards must also be flexible in the ever-changing environment. To respond to these changes and to meet these diverse demands we need global standards that would provide a platform for compatibility and interoperability of telecommunication equipment and services. It is perhaps the de facto regional organization though Australia and Singapore are becoming increasingly active in telecommunications. one every 18 months on the average. 1994: p. These are made largely feasible because of liberalization. intelligent networks. and the increasing desire for high speed access. The US is also a member of the North American regional organization on Telecommunications. thereby increasing sales and achieving economies of scale. The telecommunications industry is changing rapidly with the widespread installation of fibre optics. EDH (Electronic Data Handling). many corporations and some individuals dedicated to working on standards. and the globalization of these industries. They produce many standards. . as well as mobile and personal communications. and is discussed in the case study for this chapter. International standards cost millions of dollars every year. Also important. 49) TTC in JAPAN TTC stands for Telecommunications Technology Committee. It is organized with a Board of Directors.11 Standardization process in Japan (adapted from Iida. The telecommunications environment is changing because of the large proliferation of PCs. privatization and deregulation of the computing and telecommunications industry in some countries. the popularity of email and multimedia. but some can taken up to 10 years in some cases. nor too late lest vested interests of equipment manufacturers and carriers become too entrenched for them to compromise and cause them to resist their products becoming outdated and replaced by products following international standards. 27 in all in 1993. enabling them to offer a more competitive and better product for a lower price. now the ITU. The coordination of this international effort is done by international organizations line the ISO and the CCITT.Telecommunications and networks Draft Standards Review and analysis of draft standards Advanced briefings of draft standards Study of advanced briefing (3 weeks) Counter-proposals Analysis Voting Figure 11. UK and the US. there are many beneficiaries. digitization of networks. T1 (Telecommunications 1).11. Each committee and the five TCS has a set of between two and eight working groups each. The TTC is the dominant organization in the Pacific and the Far East. Business can expand their markets abroad. B-ISDN.

This contribution is considerable for it is estimated that over 200 000 impressions are made for a normal meeting. IEEE Communications Magazine. 52 55. T1E1 is concerned with physical transmission interface. There are also interrelationships with international organizations such as with the international organization ITU (successor of CCITT) through the US National Committee which is coordinated by the Department of State (Foreign Ministry). they tend to deal with short-term considerations. ‘Standards by consensus’. They agree to temporary alliances for specific projects among national and international competitors for they all stand to gain from the participation with peers. The effort on the B-ISDN is contributed by over 200 individuals coming from more than 60 organizations. Self interests are often at stake. industry and trade associations. especially for the rural and remote areas all around the world. The ultimate beneficiaries are the public at large who benefit from the many applications that telecommunications made possible. and of making the parties concerned participate in the development of international standards.Standards They are slow in delivery because of the inherently slow process of downstream and upstream feedback and consensus building between corporations. The Forum is The Forum focuses primarily on customer premises interests and produces a consensusbased standard. there are still some unresolved matters that relate to international standards in telecommunications. are quite important to mobilize many segments to develop actual products and services based on these technologies. T1M1 Focuses on network management issues. 1994: p. as well as national and regional bodies that often have vested commercial interests if not national interests to support. 121 . there evolved another layer to the traditional standards bodies called the Forum. Source: Amy. patents and experience. This raises the delicate problems of extracting proprietary information for sharing. and T1S1 in concerned with network signalling protocols and network service definition. Because these agreements are driven by the impetus of commercial products and services. T1 is the North American standards organization. Members who have voting rights pay a membership fee and every sponsoring organizations pays for the expenses and the time allocated to the project. Despite this laudable effort.1: Development of international standards for the B-ISDN in the US The development of international standards in the US is different from that in other countries largely because it is greatly influenced by its experience in the development of the 802 standards by the IEEE (Institute of Electrical and Electronic Engineers) and the development of the TCP/IP (Transmission Control Protocol/Internet Protocol) standards by IETF (Internet Engineering Task Force). however. (1994). This was true of the Forum on the ATM (Asynchronous Transfer Mode) in the development of consensus on standards for the B-ISDN standards. This is covered more fully in Part 3 of the book. and that if they must change then the change should be easy and inexpensive. Manufacturers of equipment want their designs to be internationally accepted thereby increasing their market share. 53). It has many working groups. The Forum along with IETF interact with the four technical subcommittees which are accredited to the ANSI (American National Standards Institute) via their parent committees. despite the actual (or perceived) loss for those who hold the proprietary knowledge. like the T1A1 group which focuses on performance issues. whilst carriers do not want to change what they already have any more than necessary. Case 11. (Amy. The Forum differs with the ANSI committees in that the ATM focuses primarily on the interests of the customer premises whilst the ANSI committees emphasize the interests of the public networks. The new group then builds rapidly on the base to reach an agreement among commercial interests on the details of implementation. 32(1). a consortium representing a very wide spectrum of industrial players highly focused on working on a tentative standard developed by the traditional standards organizations. One is the transfer of proprietary knowledge and technology to the public domain. Robert M. They must be brought into the consensus-building and cooperation process of developing international standards. Such agreements. Encouraged by the success of the IETF and the IEEE 802 standards.

46 49. . Shaping Future telecommunications: the challenge of global standardization. 20(9). 30(7). and Thomas. The TMN solution has its critics including the spokesperson of the German giant Siemens for having ‘always been divided into four layers element. This certainly does not facilitate the rapid provision of comprehensive solutions because the standard interfaces are required to link the individual layers. Standardization on standards. IEEE Communications Magazine. Peterson. and Sparell. 12(5). Nak. 25(1). IEEE Communications Magazine. (1995).’ Working against these criticisms of the TMN solution is the ETSI which has helped to flesh out the TMN framework along with customer administration and fault and traffic management. Do you really need ISO 9000? Open Computing. 32(1). (1992). I. Mossotto. the Network Management Forum. 65 66. IEEE Communications Magazine. the European Telecommunications Standards Institute. 1(4). Electronic data interchange: a new frontier for global standards policy. Universal Communications System.H. service and business management. IEEE Communications Magazine. Telecommunications Management Network. Tristam. C. (1991). and the Tina-C. 22 28. 122 . Standards and innovation in telecommunications. broadband and access management. Pathways for telecommunications: a European look. 54 61. the NMF. Intelligent Networks. The ITU-T is mainly concerned with TMN. NMF is working on information module definitions that are available for transmission. 12. p. (1992). International standards for intelligent networks. (1994).S. D. C. C. Trauth. Byte. 26(9). Coordinating global standards and market demands. (1995). 11(7). the ETSI. 72 75. Mostafa. 11. . 607 619. and Dvorak. 32(1). network. 32(1). J. 1995. the International Telecommunications Telecommunications sector. These are: ITU-T. (1993). Telecommunications standards development.M. E. 30(2). Domestic standards in a changing world.’ Source: International Herald Tribune. R. 68 70. 6 17. F.K.A. (1992). (1994). Global standards. IEEE Communications Magazine.M. switching. 52 57. G. Computers and Security. 20 28. Telecommunications. 38 42.2: Networking standards in Europe There are four organizations in Europe that are concerned with standards of networking. 32(1). H. Mazda. Oct. Telecommunications. Standard issue. Bibliography Burton. and Visser. (1992). the Telecommunications Intelligent Network Architecture Committee. Duran. (1994). ETSI has also made recommendations to the ITU on UPT. IEEE Communication Magazine. 34 36. Iida. T. J. It attempts to increase the increase the IQ of the IN ‘with focus on distributed processing and object communications . D. T. 26(8). J. (1992). 201 205. J. Knight. LAN security standards. Irmer. Journal of Global Information Management. David. (1993). Tina-C embraces the TMN as well as IN. IEEE Communications Magazine. also addresses service management more specifically than TMN.Telecommunications and networks Case 11.

. because these considerations are common to corporate security (and not entirely unique to telecommunications). The one exception is the assessment of risk and the determination of how much security is needed. 90 91). The opportunity is great and ‘opportunity makes a thief’. Anyone who can scrounge up a computer. a long distance telephone company cardholder compromised his card and 600 unauthorized international calls were placed on that card before the network specialists detected the problem and disconnected the violator. In European countries such as the Netherlands. for instance. We need to use technology and devise security policies that not only catch the violator but dissuade the violator from breaking into the system. We have to be faster and smarter. 1994: p. Hence these considerations lie outside the scope of this book. We will not discuss organizational considerations of security such as administrative controls. assign their students sites on the Internet to break into and files to bring back as proof that they understand the protocols involved. . 123 . In it we examine the threats and responses to access which is a new twist to an old problem of computer centre security. Computer science professors . In one case in 1993. computer intrusion is not necessarily a crime . the appointment of a security officer and operational security. .12 SECURITY FOR TELECOMMUNICATION Every new technology carries with it the opportunity to invent a new crime. All this happened in less than 2 minutes. (Wallich. because this is somewhat different and often more critical in telecommunications. This increased the points of vulnerability to intrusion and unauthorized access. A study for the period of 1964 73 identified 148 cases of computer crime that were made public.5 billion. There are also many people who are either using computer systems or have access to computer systems. 1994: p. industry numbers range from $1 billion to $9 billion. . One author has identified 55 sources for potential computer intrusion. . The US Secret Service has estimated the annual cost of fraud by telecommunications at around $2. It is such technologies and policies that are the subject for this chapter. . We also examine the unique threats to telecommunications including the threat of computer viruses. . a modem and $20 a month in connection fees can have a direct link to the Internet and be subject to break-ins or launch attacks on others . Telecommunications has only increased the exposure to computer crime. With every change in technology comes an opportunity to violate the system and the newer intrusions are always getting to be more creative and effective. some of the most respected domains on the Internet contain computers that are effectively wide open to all-comers the equivalent of a car left unattended with the engine running . Some intrusions are the unintended consequences of corporate action when in their enthusiasm for downsizing. (Wallich. 94). . Laurence Urgenson Introduction Problems of computer security go back to days long before telecommunications and LANs became ubiquitous. they pushed applications (and sometimes mission critical applications) on to remote servers which had to use a LAN or some other telecommunications system. We cannot continue the traditional approach of ‘security through obscurity’ which is the keeping of vulnerable data secret. One reason for this high cost is that PCs are now very common and the population that know how to use PCs is so large.

With on-line systems using telecommunications. guard dogs and security check stations are procedures common to restricted areas of manufacturing plants and government installations where work with secret or classified materials takes place. flood. and regulate personnel. The very high volume of business information processed by computers today means that the rewards of industrial espionage and fraud are of a much higher magnitude than in the past and are ever increasing. electric fences).1. Large sums of money transferred daily by electronic fund transfer must be protected against theft. Security measures described below are designed to guard information systems from all the 124 above threats. Other measures guard against physical plant. window bars. physical barriers (locked doors. security is a greater problem. pillage or unauthorized use of the computer itself. Confidential data on market strategies and product development must be kept from the eyes of competitors.Telecommunications and networks Legislative Personnel Operational Communications Authorization Terminal use Inadvertent human mistakes in Programming Design Input Output Procedures Operations Plant Invasion of privacy Danger of natural disasters and accidents deterrents policies and screening security hardware/software controls security Espionage fraud and theft Figure 12. capture. A vault for storage of files and programs and a librarian responsible for their checkout provide additional control. since stringent access controls to terminals may not . hurricanes. Some controls guard against infiltration for purposes of data manipulation. a breakdown in air-conditioning may cause some computers to overheat. Records must also be protected from accidents and natural disasters. a buffer zone. These measures can be envisioned as providing layers of protection as shown in Figure 12. For example. These controls are now discussed below. resulting in a loss of computing facilities. monitor operations and telecommunications. alteration of computer programs. and even a heavy snowfall causing a roof to collapse can cause the destruction of data and valuable computer equipment. Personal data are not the only vulnerable data. Terminal use controls Badge systems. Fire.1 Layers of control Security Security is where data (and information) are protected against unauthorized modification. and destruction or disclosure.

Security for telecommunication exist at remote sites. When given a choice. or they tape the paper to the terminal itself. such as infrared. A biometric system can also be based on voiceprints. For example. pattern recognition systems are still not problem free. then converted into digital from which a set of measurements are derived that identify the voice pattern of each speaker. as determined by some biometric measure or physical characteristic. such as a password. One example of a password system is the required use of a personal identification number to gain access to an automated teller machine at a bank. The disadvantage of both keys and cards is that they can be lost. where personal identification codes can be stored. Health or mood that changes one’s voice can prevent a voiceprint match. Someone determined to access the computer will make guesses. Many have difficulty recognizing patterns under less than optimal conditions. such as voice plus hand analysers. For example. an electronic scan may be made of the hand of the person requesting terminal access. Identification can be based on: ž what the user has. their possession does not absolutely identify the holder as an authorized system user. users frequently select a password that they can easily remember. In addition. In this case. with a strip of magnetically encoded data on the front or back. Such systems are based on the dynamics of pen motion related to time when the signer writes with a wired pen or on a sensitized pad. Some have a core of magnetized spots of encoded data. ž what the user knows. wives or children. therefore. house number or names of pets. Biometric control systems. have been under development for many years. cut. Passwords The use of passwords is one of the more popular methods of restricting terminal access. A microprocessor in the reader makes an accept or reject decision based on the card. Optical cards encode data as a pattern of light spots that can be ‘read’ or illuminated by specific light sources. In other words. security systems use electro-optical recognition and file matching of fingerprint or palm print minutiae. Signature verification of the person wishing to log onto the computer is yet another security option. ascertain the identity of persons who wish to log on and must determine whether they are entitled to use the system. can interfere with a fingerprint match. trying such obvious passwords 125 Biometric systems Some terminal control systems base identification on the physical attributes of system users. the use of passwords is often an added security feature of key and card systems. identification depends on matching: the voice pattern of the person wishing computer access is compared with voice profiles in computer memory. The computer itself must. a voice profile of each authorized user is recorded as an analogue signal. The problem with passwords is that they are subject to careless handling by users. A combination of devices. For this reason. the reader for this card must include a transmitter and receiver. of special interest to defence industries and the police. Keys and cards Locks on terminals that require a key before they can be operated are one way to restrict access to a computer. similar to credit cards. Another way is to require users to carry a card identifier that is inserted in a card reader when they want to use the computer. Only a positive match will permit system access. Proximity cards contain electronic circuitry sandwiched in the card. ž who the user is. ‘God’. Some use plastic cards. might ensure positive identification. Such . even sweat on hands. ‘Pass’ and ‘Genius’. such as an ID card or key. a blister. and microprocessor intelligence. Again. such as their birth date. The chip has both coded memory. Many types of card system are on the market. Top of the list in Britain seems to be ‘Fred’. This scan is then measured and compared by computer to scans previously made of authorized system users and stored in the computer’s memory. Some users write the code on a sheet of paper that they carry in their wallet. stolen or counterfeited. Fingerprints or palm prints can likewise be used to identify bona fide system users. Although technological breakthroughs that enable discrimination of complex patterns have been made recently. inflammation. there are smart ID cards that have an integrated circuit chip embedded in the plastic. but such equipment is too expensive at the present time to be cost effective for most operations in business.

usually youths. Typically. The data elements accessible from each terminal can likewise be regulated.00 08. A sample printout from an access director.1. a challenge when portable terminals are used by personnel in remote sites where security may be lax. The computer responds with a ‘challenge number’. But systems of this nature are difficult to administer. Recently. This compounds the problems of a person trying to breach the system in a network Table 12. Authorization controls In addition to the identification systems outlined in the preceding sections. usually a handheld device. In recent years. Information is power.00 08. the terminal in the database administrator’s office might be the only terminal permitted access to all files and programs and the only terminal with access to the security matrix itself. One-time passwords are a viable alternative.00 12. This is done with a central intelligent controller at the host site and a random password generator for each user. Finally. and easy to use. Assigning access levels to individuals within an organization can be a difficult task. write or update). a one-time password is generated. the computer might be programmed to reference a table that specific the type of access permitted or the time of day when access is permitted. and the right to access it is 126 . Data directory A computer can be programmed to reference a stored data directory security matrix to determine the security code needed to access specific data elements in files before processing a user’s job. Then there must be agreement on the method of selecting the next valid password from the list. The advantage to the user is that the password generator is portable. Furthermore. and to ascertain what type of access is permitted (read. the longer a password is in use.00 17. because passwords are constantly changed. Of course. First of all. a method that is synchronized between computer and user. When the user lacks the proper security clearance. the control system is protocol dependent. Even passwords as complex as algebraic transformations of a random number generated by the computer have been broken with the assistance of readily available microcomputers. storage of the list must be secure.00 having a variety of protocols. a number of password systems have been put on the market that generate a new password unique to each user each time access is attempted. who often derive malicious pleasure in circumventing computer access controls. control systems can be installed to verify whether a user is authorized to access files and databases. the system works as follows: To gain mainframe access. access will be denied.00 17.1 Access directory User identification: 076-835-5623 Access limitation: 13 hours (of CPU time for current fiscal year) Account Number: AS5842 Data Elements Customer number Invoice number Cash receipt Type of access Read Read Read/write Security level 10 10 12 Terminal number 04 04 06 Time lock 08. each authorized user must be given a list of randomly selected passwords. In a similar manner. the user enters his or her name (or ID code) on a terminal keyboard. sorted by user identification number. the greater the likelihood of its being compromised. according to a programmed rule. The user then enters this password into the computer.Telecommunications and networks first. For example. Only a short period of time is allowed for entry of the correct password. is shown in Table 12. By applying a cryptographic algorithm and a secret key (a set of data unique to each password generator) to this challenge ‘seed’. Such password management systems are difficult to compromise. This is input to the user’s password generator. The central controller simultaneously calculates the correct password and will grant access if a match occurs. much publicity has been given to hackers.

Although a number of projects have attempted to demonstrate the practicality of this security approach. Figure 12. Figure 12. a systems component that checks each reference by a subject (user or program) to each object (file. In a multiuser system. It is important that the need for security be understood by workers and that security controls be administrated with tact. thereby circumventing system security.2 Raiding files: a Trojan horse program Program A Smith Reference monitor Data file A Program B (game program with embedded Trojan horse) Brown A fil cce bl e A ss oc to ke d Figure 12. Although the data directory does not authorize Brown to access File A. A security kernel represents new technology still in the developmental stage. Employees may vie for clearance even when they do not require such clearance for their jobs. data in a file can be raided by installing a ‘Trojan horse’ program.3 How a reference monitor blocks a Trojan horse raid es us B ith ram Sm rog p Data file B 127 . Manager should recognize that security measures designed to protect confidential data and valuable computing resources may antagonize loyal employees. A kernel is a hardware/software mechanism that implements a reference monitor.2 shows how this is done. results thus far have been mixed. data from that file is copied into another file that Brown is entitled to access. Data File A (sensitive data) Security kernel Unfortunately.Security for telecommunication a status symbol. device or program) and determines whether the access is valid according to the system’s security policy. the use of a security matrix does not provide foolproof security. The concept of a security kernel addresses the Trojan horse issue. confidential Program A Pr o co gra pi m e fro s d sec at re m a tly in File Fi A le B Security matrix File A B Data File B B Authorized user Smith Brown Smith Type of access Read/ write Read/ write Write es us B ith ram Sm rog P Program B (game program with embedded Trojan horse) Smith Brown Figure 12. on the direction of a secret program.3 shows how Brown is foiled by a reference monitor in his attempt to raid File A.

In transposition. a business looks up Firm X’s E key. Firm X alone has the secret D key for decryption. This system has two keys. a key that both sender and receiver possess. Protecting the confidentiality of this data at the initiating terminal. most schemas are highly complex. from the Greek root ‘crypt’ meaning 128 to hide.Telecommunications and networks Virtual machine An entirely different approach to security in a multiuser environment is a virtual machine. This may be eavesdropping. so that 514 reads 534. or encrypt. With the increased reliance of businesses on . such as piggybacking (the selective interception. Each sender/receiver has a set of D and E keys. there is no longer need for secrecy once a deal is made. and privacy (impedes eavesdropping). such as a 2 added to the third digit. and a D secret decryption key used by the receiver. Understanding them may require mathematical knowledge and technical expertise. an E public encryption key used by the sender. the third and fourth characters might be switched so that 5289 becomes 5298. making 514 read 516. the key must be difficult to break. conventions. control (prevents alteration of a message). can be done by either transposition or substitution.4. characters are exchanged by a set of rules. this isolates one user from another although they use the same physical machine. As in all codes. modification or substitution of messages). procedures for user identification (described earlier in this chapter) and dialogue termination also help maintain the confidentiality of data. A specified number might be added to a digit. characters are replaced. during transmission itself or when transmission is received. Communications security Computer processing is today closely linked with telecommunications. pretending to be legitimate users of the system. A key is used to code the message. An illustration of encryption appears in Figure 12. for example. This prevents individuals from masquerading. because each virtual machine can be operated at a separate security level. In substitution. Or the substitution may be more complex. the electro-magnetic pickup of messages on communication lines. is one way to control communications. a predetermined signal that the computer must recognize before initiating transmission. Most companies use callback boxes that phone would-be users at preauthorized number to verify the access request before allowing the user to log on. An additional problem is that there sometimes is insufficient time to pass the key to a legitimate receiver. In effect. One method of preventing message interception is to encode. One solution is a multiple-access cipher in a public key cryptosystem. It could be a random-number key or a key based on a formula or algorithm. This and other uses of unauthorized time can be quite costly to a business firm. Decryption restores the data to its original value. has required the development of sophisticated security techniques. or a patent application is filed. With a virtual memory structure. messages are vulnerable to wiretapping. For example. In the past. the stock market closed. because the value of much data resides in timeliness. Frequent changing of the key adds to the security of data. An illicit user taps the computer when a bona fide user is connected to the system and is paying for computer time but is ‘thinking’. published in a public directory. For example. in effect. This system can be breached but not easily. Often. With this systems structure. Although the principles of encryption are relatively simple. since a tremendous number of computations would be needed to derive the secret of D. data in order to render it incomprehensible or useless if intercepted. and then transmits a message in code over a public or insecure transmission line. so the computer is idle. passive listening or active wiretapping involving alteration of data. A hacker who has learned the handshake code would be denied access with such a system. Protocols. The code’s security lies as much in the time required to crack the algorithm as in the computational complexity of the cipher. a handshake. each user loads and runs his or her own copy of an operating system. During transmission. Another type of infiltration is reading between the lines. which explains why many systems have a key base with a large number of alternate keys. which allows the transference of computer data between remote points. several user programs can reside in computer memory simultaneously without interference. the transportation of the encryption key to authorized users has been as Achilles’ heel to systems security. Encryption. The number 1 may become a 3. Cryptography. To code data to send to Firm X. serves three purposes: identification (helps identify bona fide senders and receivers).

Authenticity of images stored in documents. Space limitations do not allow us to examine all or even some of them in any detail except perhaps one: image processing. PGP was written by a programmer in Colorado who has made it available on the Internet and some non-Internet bulletin boards for no charge. much research has been done on cryptographic systems. 129 . Integrity of images and unauthorized modifications made to them. the document would turn into gibberish when put on a bulletin board or transmitted via e-mail. However. These threats include: ž ž ž ž Unauthorized copying and downloading of images on terminals. PCs and workstations. This has angered the US Government and frightened its agencies of law enforcement. The hardware implementation can keep up with high speeds even with gigabits of high speeds.Security for telecommunication Sender Message sent (plain text) Encryption Message transmitted Decryption Receiver Message received (plain text) Key base Key Key Key base Example (in binary numbers) Message− original input key− to add Encrypted message− (output) 1 1 0 0 Encrypted message− (input) key− (subtracted) Original message− (output) 0 1 1 0 1 0 1 0 1 0 1 0 0 1 1 0 1 1 0 0 Figure 12. This protection against unauthorized transmission is attractive since encryption algorithms protecting privacy of documents can be protected by codes such as the PGP (Pretty Good Privacy). Unauthorized release of images by users (including end-users). speed and advanced technology of modern computers. thereby losing the US a lot of export. it has been argued that this may be a violation of freedom of speech and expression. One solution to this problem is to embed a cipher in a document identifying both the owner and the user. If a person purchased one-time rights. using the power. Some claim that persons bent on accessing encrypted data will be able to break all codes. terrorism and computer crime. This will illustrate the types of threats to security posed by one advanced technology. Also. In contrast.2.4 Encrypting and decrypting data in teleprocessing teleprocessing. The US government wants this code to be part of all encryption software that is exported. computer technology is well known for its many entries (and exits) of innovation. But hardware implementations are more expensive in initial costs as well as in space occupied and running costs of power consumption. Cryptography can be implemented through hardware or through software. Some of these are listed in Table 12. US exporters argue that foreigners will not buy such a code and will get encryption codes elsewhere. But experts disagree about how secure even the most complex codes are. defense and intelligence. And so the controversy rages. Encryption can be mathematically complex and yet it can be decrypted given enough time and motivation. we have discussed the security considerations that are traditional to IT. software implementations may not always keep up with high speeds but are more versatile with solutions such as encryption. Security for advanced technology Thus far. They support the ‘keyescrow encryption’ which allows the government to hold keys to any encrypted communication to be used for occasions of national security and the fight against money laundering. The user is the buyer who can purchase a cryptographic key to decode the document.

file on the system. . as well as a cost of implementation of a response to the threat.2 Technologies affecting security Teleprocessing LANs/MANS/WANS/Internet Superhighway Global networks PCN (Personal Computer Networks) Wire-less communications Fax connections Image processing Smart cards NNets (Neural Networks) Laptop computers Palmtop computers Pen-based computers PDA (Personal Digital Assistant) Personal communicators Pocket pals Teleconferencing Video-conferencing Computer viruses A computer virus is software that. which is a computer program that seems to do one thing but also does something else. All this unbeknown to the user . Also. and potential losses. Another sort of computer vermin is the worm which is a program that ‘worms’ its way through a system. The virus can act instantly or lay dormant in the system till it is triggered by a specified date or an event such as the processing of five programs. It is definitely unauthorized. it is inserted (knowingly and deliberately) into the system. 26) describes the Lehigh virus (named after the Lehigh University in the US). management must maintain knowledge of the technological changes and its potential impact on security. A virus is a class of programs. How does a virus operate? Zajac (1990: p. It operated as follows: . One way a computer virus can enter the system is in the form of a Trojan horse. and develop protective defense measures necessary to combat the threat. It could also corrupt data. . .Telecommunications and networks Table 12. However. or the logging on as a specific user. it would totally erase the hard disk. when entered into a computer system can cause it to stop or interrupt operations. destroy data/knowledge-bases. when a user typed a DOS command. exposure to risk. Some cases illustrating its variety are listed below. To achieve this. 130 The computer virus has many objectives and can have many consequences. . it uses minis and mainframes which are accessible through a network and telecommunications. which is a computer program that is set to ‘explode’ when a specified condition is met. . the virus has a new and malicious twist to it. or whatever. The computer virus has the ability to propagate into other programs. there is the logic bomb. it infected it and incremented a counter that kept track of how many other disks it had infected. If so. . . Management must continuously evaluate the consequences of failure to protect its assets of information and the cost of updating its protection against technological changes. Since image processing requires large memory and fast computing power. and cause errors or disrupt operations. the virus would check to see if there was a noninfected . altering small bits of code or data when it gets access to it. Continued violations of computer and telecommunications security should be anticipated especially from the computer virus. For this to happen. In contrast. This virus consisted of seven lines of code in Pascal and was placed in a DOS command file. A worm may also be a virus if it reproduces itself and infects other programs. a computer virus must be introduced into the system. The virus would then execute the user’s command. This opens a wide set of threats. the computer virus program must be run in order to reproduce or to do damage. But.2 present a unique set of benefits. anticipate threats and vulnerabilities. the danger increasing with telecommunications and networking. The virus does not just appear or grow. when the infection counter hit four. Each of the technologies listed in Table 12. The anticipation aspect is important since the protective measures must be instituted in the design stage of development and sometimes even earlier in the planning stage and user specification stage of development. the large computers necessary for image processing contribute to the temptation of using these powerful resources to break the security codes of other systems.

the ‘Bulgarian factory’ replaced Israel as the source of viruses and over 100 viruses from Bulgaria have infected the Eastern European countries. She then used this program on the firm’s network thereby infecting the network and erasing many corporate files. It cruises stealthily inside a system for a long time before it strikes. ‘Once inside the system. each disk was infected with a STONED II computer virus. Unknown to the publisher. Inadvertently. 38% were confronted with corrupted files. The reward for this type of intruder is a rise in ego. interference or lockup.’ The message was followed by a 1986 copyright date and two names (Amjad and Basit) and an address in Pakistan. The main motivations are greed for money. and 62% reported a loss in productivity (Sanford. a virus appeared in a computer network in California and interfered with the scan control on video monitors and caused one to explode. The main counter-strategy for (1) is to: ž ž ž control access to programs and data/knowledge and susceptible media invest in people who are the greatest threat and also the greatest asset to security for they can often stop or at least discourage intruders monitor and control access to all ‘vital’ computer programs as well as data/knowledge even when provided by insiders especially those that may be disgruntled or unhappy with the organization 131 . the V800 and the V2000. This naming is perhaps because the authors worked for ‘Brain Computer Services’. When loaded onto a system it collects information like names and passwords and can be accessed by the intruder and used for authorized access to the system. An employee in a large firm used a program (not known as being infected) available externally on a public network and downloaded it to her PC. The BRAIN virus surfaced in the US at the University of Delaware in 1987 followed a month later by the Lehigh virus. This type of intruder is a professional who does not get the public media attention but accomplishes a sinister mission none the less. How can an end-user protect a computer system from these infectious and dangerous viruses? One answer lies in knowing the possible motivations and the sources of the viruses. 140). The virus is also known by its generic type: the BRAIN virus. Over 60% of the 600 000 PCs studied had been hit by a virus. named after the volume label of an infected disk which reads ‘(c)BRAIN’. 67). Contact us for vaccination. One study in 1991 estimated that there are over 900 known viruses.Security for telecommunication The Pakistani virus infected untold number of PCs as it travelled around the world creating havoc and fear among PC users. One may ask: How many viruses are around? The answer can be found in various studies. She used the disk (which was infected and unknown to her) to do her homework assignments. the intruder unleashes the virus. The employee is motivated by greed or by revenge. 1993: p. In the early 1990s. It is known to be designed to elude most of the anti-virus systems. The ‘stealth’ virus is named after the stealth bomber that attacked Iraq in the Gulf War. In 1987. and the ‘intellectual challenge’ to outsmart others. she infected all the students using the network and caused the system to operate incorrectly. Its versions include the 4096. revenge (against an employer or a firm). 41% complained of unsolicited screen messages. The Cornell virus was a passive virus with the intent of collecting names and passwords. A student in California was given a disk with a free program. Of the sites infected. She took the disk to her university and loaded it on the university network to continue with her assignments. The Jerusalem virus followed the Delaware virus by two months and its first strain appeared in the Hebrew University in Israel. The intruder that uses the open distribution channels is mainly motivated by the intellectual challenge of breaking a system. which lurks there until an authorized individual decrypts material or enters the access sequence. A ‘cruise’ virus is similar to the Cornell virus. It is a sophisticated passive virus that could infect disks that are distributed openly. There are two main sources of computer viruses: (1) an employee (insider) and (2) openly distributed programs such as those distributed by magazine publishers and software vendors.’ (Dehaven. the virus then attacks the material. 1993: p. The authors of this virus had the gall (or courtesy!) to announce its existence with the message: ‘Beware of this VIRUS. A magazine publisher in Germany distributed over 100 000 disks to its subscribers. It infected thousands of mainframe computers throughout the world to enable them to be used later as deemed desirable.

Telecommunications and networks ž control all access during the conversion phase of new systems. responsibilities and accountability to those who own and those who manage the messages being transmitted ž specify who should have what resources and for what purpose ž devolve responsibility (whenever possible) to the organizational point where controls are implemented ž ensure interoperability ž ensure basis for security review and audit ž base security access levels on risk involved so that there is no or little unreasonable imposition ž ensure sharing of network in a responsible and controlled manner ž protect resources from misuse. There is also a loss in efficiency. The anti-viral strategies we 132 have are against ‘known’ viruses only. unauthorized use. there may be a need for a set of policies which can be carried out modularly in increasing . A strategy to prevent this is for a copy of the tested system to be locked-up and used periodically to check for any unauthorized insertions. the next subject of our discussion. scanners and filters (programs that check for ‘signatures’ of known viruses and alert the user to a possible danger) ž keep abreast with the virus technology one journal on the subject is The Virus Bulletin published in the UK. There is a possible loss of morale when employees are not fully trusted. In addition to calculating these costs. ž control access to networks. This logging may not always identify the intruder but it may dissuade the intruder to regularly change the common systems passwords ž educate and instruct employees of the danger of viruses and their epidemics keep track of where your disks/tapes and programs (including updates) have come from and where they have been Controls for both (1) and (2) are: ž latest and frequent back-ups to recover. Do not allow persons or ‘workstations’ access without the ‘need’ to such access. especially the ‘intellectually motivated’ intruder. may be challenged by the control mechanism into finding a new strain and a new twist to an old threat or devise a new threat to beat the systems that we have not yet heard of or even thought of. To carry out these objectives for security. one must estimate the probability of attack and the value of the loss entailed if the attack is successful. When access is allowed it is logged. This analysis is necessary before a security system is designed and implemented to combat viruses. There is also a cost to all this control and strategies against viruses. and prevent unauthorized traffic (as defined by network policies) from inside to outside or the other way around. A combination of filters (also called screens) and gateway(s) act like a wall against viruses or other unauthorized traffic. To guard against viruses and other threats to security. malicious actions and carelessness. a firm needs policies that will help make the system secure. The main danger of all these strategies is that they may lull the potential victims into a sense of being protected. They are called firewalls. The objectives of these policies could be: ž specify security measures and specify and assign roles. Corporate managers and end-users must recognize that the intruder. Policies for security To implement and enforce any set of security measures there is a need for policies. This is when the system has been satisfactorily tested and every one relaxes. sometimes an extra generation deep ž anti-viral computer programs. The important controls for (2) are: ž do not use unknown software ž mainframes and even PCs should have a ‘quarantine box’ to sample check new software ž centralize software purchasing and purchase only from an approved list of vendors ž do not use freely distributed disks unless they are ‘reliable’ and tested in a ‘quarantined’ program. a perfect time for the intruder to sneaking a virus into the systems. Each layer and level of security has an overhead cost and loss in productivity and performance.

Security for telecommunication progression of complexity. Connection rules These specify the operational implementation of the broad overall network security policy. cards or passwords ž set up and manage accounts for users ž check for viruses ž check for backup and integrity of telecommunications system ž set up logging procedures ž set up monitoring of successful and unsuccessful attempts at accessing system. These policies (Symonds 1994: p.5. The algorithm will then Request for authentication Information on subject AUTHORIZATION Information on object ALGORITHM Authorization authority Analysis No OK ? Authorization denied Authorization approved Yes Access type information Figure 12. dispatch and delivery of messages ž definition of authorized origin ž definition of what constitutes ‘authorized’ traffic ž definition of responsibility of custodian and ownership of messages ž definition of the ‘rights’ to access by a remote user ž acceptance of compliance of standards for telecommunication accepted by the organization ž responsibilities as to the protection and security of messages. MODIFY or DELETE. 479 80) can be: Access agreements These are ‘agreements’ hopefully mutually agreed upon but if necessarily imposed that specify: ž definition of origin. Network security policy This needs to be an overall policy stating such principles as: the responsibility for network security lies with the individual managers of the end system or subsystem except when central control is needed for the benefit of all the users. the subjective variables include the role and record of the person seeking authorization. Domains for security control and management are assigned with specification of necessary authority and accountability. Here the objective attributes are such considerations as the importance and sensitivity of the authorization involved. These rules include: ž rules for construction of access media whether this be badges.5 Process of authorization 133 . or instituted all at once as a comprehensive security plan. READ. such as WRITE. Administration of authorization Authorization can be the consequence of a formal computation of values of authorization variables as shown in Figure 12. and the access requested will involve the type of access desired. destination. ž set up displays for warnings of possible misuse or abuse of system.

in part.8 8 000 Total £168 000 expected loss One decision not determined directly by policies but equally important is the determination of how much security is needed. 0. An algorithm may also be used to determine the level and type of authorization. In addition to the expense of equipment and personnel to safeguard computing resources.5 60 000 3 50 000 0. 4. firms are reluctant to publicize how their security has .Telecommunications and networks determine whether a request for authorization should be granted or rejected. but a lot of unnecessary bad feeling can also be generated by large egos and sensitive people. If PA and PB are probabilities for the year. where more than one person makes the decision.2. Opportunities for systems invasion. centralized. One way to calculate expected losses from intrusion is by application of the formula: Expected loss D L ð PA ð PB .0 0. This is our next and last topic for this chapter. The resources available to deter or counter a security breach should also be appraised. Rather than attempting to assign a specific value (0. Probability values are more difficult to obtain. expected loss is £168 000 per year. the sum of the expected losses is the total risk value to the system.6 systems and user characteristics should be assessed when evaluating risk. We simply do not have the data to calculate reliable probabilities. This formalization is desirable in order to move the decision-making from a subjective and hence appealable decision to one that has the perception of being objective. where the authorization is assigned by the owner.2 0. PB D probability of success. The application of this formula will help management determine whether the addition of security measures is worth the cost and where the greatest reduction of expected losses could occur by improving security. As illustrated in Figure 12.05 may be of spurious accuracy). fraud. motives of a possible invader. where L D potential loss. 5. medium or low) should first be determined and a numerical value assigned to each of these relative probabilities (for example. How much security? Security is costly. 2. delayed processing. Furthermore. The risk costs can now be calculated according to the formula. on evaluation of expected losses should the systems be breached.045 or even 0. cooperative. PA D probability of attack. management should analyse risk. How exposed and vulnerable are the systems to physical damage. then not only can time and energy be consumed in appealing the decision. other costs must be considered. In deciding how much security is needed. where a higher level manager delegates the authority to a lower level person who then exercises all the authority. such as employee dissatisfaction and loss of morale when security precautions delay or impede operations. where one person or a group of persons is authorized to grant and revoke authorization. For example: Exposure L ð PA ð PB D Expected loss 1 £500 000 1. where the owner of the object created assigns the authority for its access. relative risk (high. If the decision is strictly subjective.2 £100 000 2 200 000 0. 0. because the computer industry is too new to have a long historical record of invasions on which to base probability assessments. The amount of security that should be given to systems should be based.8. respectively). An insurance company or computer vendor can help management determine the value of L. The administration of network security policies can be carried out in one of five ways: 1. decentralized. hierarchical. ownership. 3. but experience has shown that more of a Security Manager’s time is spent in personal matters than one would expect. This authorization process may seem like a lot of bureaucracy for a simple operational matter. disclosure of data or physical threats? What threat scenarios are possible? 134 Loss is determined for each exposure. The figures derived from the formula are approximations at best. Personal emotions are often evoked on matters that involve authorization and security levels for access.6 0.5 and 0. and resources that might be allocated to invasion should be considered.

135 . Management’s dilemma is not whether security is needed but how much. is one method of helping management determine which security strategies are mot cost effective and which policies are to be revised to stay relevant despite the changing computing environment. This means data on security infractions are incomplete. and the assessment of expected losses and gains from security protection. given enough time. More serious.Security for telecommunication System characteristics Threat scenarios User characteristics Information systems Exposure and vulnerability of information system Decision to safeguard Invader's opportunity for invasion Invader's − Resources − Motivation Assignment of − Resources − Technology − Procedures Invader Safeguards Probability of − Attack − Success − Expected losses Figure 12. in part. so news of security invasions is seldom broadcast. to the temptation arising from the large sums of money being transferred by electronic fund transfer and to the fact that more criminals are becoming knowledgeable about computer technology and are equipped with powerful computers to help them plan and execute their crimes. Risk analysis. resources and ingenuity. All known protective mechanisms in telecommunications can be broken. cost and risk) and too time consuming (in planning effort and time needed to actually break safeguards) to make attempted violations worthwhile. This can be attributed. There are also individuals who are challenged simply to ‘beat the system’.6 Factors in assessing expected losses from systems intrusion been breached lest their credibility suffer. Perhaps the major objective of security systems should be to make intrusion too expensive (in equipment. Computer crime is increasing at an alarming rate. Summary and conclusions There seems to be universal agreement among knowledgeable commentators that computer crime figures are destined to rise unless the computer industry and organizations that use computers and telecommunications pay greater attention to security issues and devote more resources to the protection of information systems. persons who design security measures are not always aware of the tricks and techniques used by perpetrators of crime to break systems security and so cannot plan countermeasures.

and procedures to beat the intruder if he or she is not sufficiently dissuaded. For example. Virus infection Copyright infringement Biometric system Authentication S E C U R I T Y Figure 12. preparing risk analysis. neural nets and intelligent systems) pose new and unique problems of security in teleprocessing. In the late 1980s and the early 1990s. IT management and corporate management must take steps of appointing a security officer. Teleprocessing security is often part of security in corporate IT. The security manager must use diplomacy and much handholding to diffuse such situations if the need-to-know/want-to-know discipline is to be maintained. however. there was great concern over privacy. concern for privacy has given way to concern for viruses. Corporate managers and network managers must be increasingly conscious and aware of the security of information systems for they may be Masquerading Piggyback riding Hacker Others Brain Logic bomb Trojan horse Password Key/token Badge Authorization Counter measures Subject Object Firewalls Encryption Locks Audit Trails Place Time Type of access Worm Smart Dump Power supply interrupted Interception Disclosure of information Unauthorized access Threats Contamination of info. And we need better hardware. We need informational laws and judicial systems with sufficient punitive penalties to dissuade the potential intruder. instituting policies and procedures for security. and software. Thus security management is not just a problem of computer technology but an exercise in human resource management. This is a problem that became public in 1984 and since then has become both troublesome and complex. Also. immunizers and memory-resident activity monitors that can defend against some viruses. In the 1970s and even early 1980s.7 Summary of threats and countermeasures 136 . Furthermore. the parasites seem to grow faster than they can be identified and information systems are compromised. but the cost of refusing access is intangible. image processing. Often. the ‘want-to-know’ is more than the need-to-know and refusing such access can cause bad feelings. Threats to teleprocessing as well as possible countermeasures in technology and in organizational policies are summarized in Figure 12. and then enforcing and monitoring these safeguards. smart cards. especially when the ‘want to know’ by a supervisor is higher than the ‘need-to-know’ of one supervised. A rational decision on access may well be that access should be based on a ‘need-toknow’ basis. identifying assets to be secured.7. Despite the growing number of scanners. tele.Telecommunications and networks Advances in computer technologies (like wireless communication. the viruses are getting better and more ingenious making them difficult to identify let alone arrest. Costs and benefits of security management are often intangible. Many viruses may be in their ‘incubation’ stage and unknown to us. planning for security needs. the problem has now become a global threat with potential insertions of a virus that cruises through a globally accessed network. the costs of access can be measured.and video-conferencing.

The technique was used to gain unauthorized access to hospital records at Manhattan’s Memorial Sloan-Kettering Cancer Center. It froze the utility’s internal files at a preassigned time. wiping out all the records stored on tape. In the next four months Berferd assaulted numerous organizations on the Internet including 300 in just one night.4 A French programmer. In 1993.1: Examples of hacking AT&T security in the UK spotted a impostor. Case 12. giving the hacker an updated file of access codes. A logic bomb was placed in the Los Angeles Department of Water and Power.3: German hacker invades US defence files Suspicion that someone was wrong at the Lawrence Berkeley Laboratory was aroused when Clifford Stoll. A ‘worm’ has destroyed many a computer memory and database. whose interest was aroused when he spotted the file. then put the system into irreversible disarray just to illustrate its vulnerability and to prove to himself that he could ‘beat’ the system. on its lines in 1991 and decided to follow him. At the time Berferd was in the Netherlands where hacking is legal. This happened to Dick Streeter when his screen went blank as he was transferring a free program from a computer bulletin board into his machine. 1994: p. three friends in Milwaukee. noticed a 75 cent accounting discrepancy. 63). but also for the ‘prudent use of the information available to the company in order to protect its customers and employees’. Loss of storage can also result from a ‘Trojan horse’ also known to security specialists as ‘logic bombs’. Stoll was able to monitor the hacker’s activities and observe that he methodically invaded files in some three dozen US military complexes to sift out information on defence topics. manger of a multiuser computer system at the laboratory. The hacker was subsequently arrested by German authorities under suspicion of espionage. Sloan Kettering Cancer Center and the Los Alamos National Laboratory in New Mexico. processing and game files that were stored in Street’s machine were erased. But the hacker’s identity remained a mystery until Stoll’s girlfriend suggested setting up a trap. Nearly 900 accounting. Shoch created a worm to wriggle through large networks looking for idle computers and harnessing their power to help solve the problem of unused resources. An Oregon youth in the US used his terminal to gain access to the computer of the Department of Motor Vehicles. not just for ‘prudent’ protection of the company’s information assets. after being fired.2: Examples of malicious damage 1. The Dutch authorities did nothing about it till Berferd attacked their machines. The AT&T lawyers decided to halt the monitoring fearing that they may be accused of harbouring hackers. a fictitious file on the Strategic Defense Initiative (also known as Star Wars) was inserted by Stoll in the Lab’s computer. bringing work to a standstill. 4. Eighteen months of detective work followed in which Stoll cooperated with law enforcement officers to track down a hacker who used 75 cents of unauthorized time. The bomb exploded two years later on New Year’s day.12. word . stayed on-line long enough to be traced. (Fried. The hacker. 3. 2. 6.Security for telecommunication held legally responsible. A ‘trap-door’ collects passwords as they log on. This use of a worm is the malign mutant of the useful worm invented by John Shoch of the Xerox Corporation in California. left a logic bomb as a farewell salute in the record-keeping software that he had been working on when fired. Case 12. 5. later known as ‘Berferd’. The malign mutant now burrows holes through computer memory. Then the following message appeared: ‘Got You’. leaving huge information gaps. They were caught and admitted that the only motivation was to ‘have fun’ and to demonstrate their technical wizardry. between the ages of 15 and 22 belonged to a gang 401 (named after the telephone area code for Milwaukee in the US) penetrated three database (a bank. German authorities believed that they cracked a major ring that has been selling sensitive 137 Case 12.

Principles of Information Systems.4: Buying the silence of computer criminals The computer industry research unit in the UK reports that the practice of offering amnesties to people who break into computers and steal funds is widespread. and without parole. Robert inserted a virus in a computer network that impacted negatively on over 6000 users. The Computer Fraud Conspiracy of Silence. He even invaded the North American Defense Command computer and DEC stealing $4 million worth of software. IEEE Spectrum. He was caught and tried. Mitnick faces thousands of dollars in fines and decades in prison. See also the delightful book by Clifford Stoll (1989): The Cuckoo’s Nest: Tracking a Spy through the Maze of Computer Espionage. Source: Karen Fritzgerald (1989). prosecuted for preventing the course of justice. The quest for intruder-proof computer systems. Then. Feb. 19 June 1989. The judge ordered Mitnick to participate in a treatment program for compulsive disorders. a computational physicist. 1995. the search ended in Mitnick’s flat where he was arrested. 3. employers who make such agreements may end up in court themselves.. 27. This angered Shimomura (30) who then cooperated with the FBI in a search for Mitnick (31). the corporation keep silent on the crimes if part of the money is returned and the swindler reveals how the fraud was carried out. His first mistake was to invade and steal software from Shimomura. 18. Source: Lindsay Nicolle and Tony Collins. They tried to contact the FBI but the FBI was on its way to arrest Mitnick and had shut off their cellular phones for fear of alerting Mitnick. and International Herald Tribune. The Independent.30 a. The Case 12. p. Then Mitnick was on parole and escaped. The FBI is pushing for a harsh sentence to deter future computer criminals. Feb. He was on bail and made another unauthorized transfer and was caught again. Soon thereafter. In 1989 Mitnick was caught. He was sentenced to three years’ probation. Mitnick’s second mistake was to taunt Shimomura by mocking voice-mail messages. Mitnick had violated Well’s system in January 1995. In one such case. 24(8). p. Case 12. convicted and put into in a low security jail. Source: Ralph M. According to a member of Scotland Yard’s fraud squad. He was back violating sensitive computer systems. 66 67. BUYING JEWELS IN RUSSIA A consultant for American bank used his knowledge of password codes and transferred money to Switzerland to buy jewels in Russia.6: Miscellaneous cases using telecommunications VIRUS OFFENDER CAUGHT AND PUNISHED Robert Morris is the son of a well respected computer expert in the US. nuclear and space research information to the Soviets. 1995. at 1. a programmer who legally diverted £8 million to a Swiss Bank account gave back £7 million for a non-disclosure agreement protecting him from prosecution. Doubleday. It was later learned that Mitnick had made a typing error and accidentally destroyed the accounting records. on 16 February 1995. 628.Telecommunications and networks military. 118(8). His mistake was that he bragged about it to friends and was caught. But Well’s management did not know that and they decided that they could not survive any more of Mitnick and had to cut him off (and thereby warn him) or risk their entire business. 22 26. p.m. Stair (1992). Boyd and Fraser. Employers fear that business might be lost if customers learn that their computer security is flawed. 400 hours of community service and a fine of $10 000. Case 12. Mitnick entered a system 5000 km away in California and wiped out all the accounting records of one of Well’s subscribers. Rather than prosecute. 18 19. Source: US News & World Report. on of all days Christmas 138 . pp. Also cooperating with the FBI was the Well network of 11 000 users.5: The computer ‘bad boy’ nabbed by the FBI Ken Mitnick was long known for burrowing his way in the most secret silicon nerve centres of telephone companies and corporate computer centres. day.

H. 12. Information Systems Management. and preventive measures. 179 180.P. Computer Security Journal. 67 72. D. 37(11). C. Cadler. 49. Svigals. Internetwork security: unsafe at any node? Data Communications. It is estimated that a 1024 bit RSA key would require 3 ð 1011 MIPS-years to crack. Wallich. Authentication of mobile users. privacy protection. S. a scientist at the Bellcore Research Institute took up the gauntlet and in May 1994 won the bet and the award. Zajac. Find the right firewall. and Tsudik. 13(6). Peukett.Jr.. A. 4(4). G. Sept. 28 31. IEEE Network. In 1993. Security considerations in a network management environment. Computer Security Journal. Security Management. Computers and Security.M. 57 63. Management Today. May/June. 26 35.C. (1994).P. Computer networks. Toward a model of security for a network of computers. R. 137 142. Nash. Hafner. J. Fried. Wire pirates.J. Salamone. Samfft. (1994). Bellovin. 64 66. VII(2). Communications of the ACM. 19 25. 26(3). ACM Computing Surveys. 20(10). (1991). I. Murray. (1997).M. 139 . and Samarai.S. (1991). T. Adi Shamir and Leonard Adelman are authors and managers of the RSA public key encryption. Sanford. (1993). and Markoff. H. Abingdon. Dehaven. 8(2). W. 50 52. M. C. J. and Choi. 473 480. Information security and new technology. H & D Special. Matching risk to cost in computer file back-up strategies. (1992).C. (1993). Parsons. (1994). Internet firewalls. 39(1). 25 31. 1995. Computer Security Journal. Bibliography Bates. IEEE Communications Magazine. (1995). 107 114. IEEE Network. Simon & Schuster.1: Popular viruses It is estimated that there are some 600 viruses cruising networks. 9. (1994). J. BREAKING A SECURITY CODE Ronald Riverset. Smart cards a security assessment.R. 2(2). remedies. International Journal of Computer Application in Technology. (1994). J. J. Security in distributed and client/server systems a management view. 61 66. 211 254. The breaking of the code has not exposed telecommunications to every hacker or computer criminal because the encryption code has been enhanced by using 512 to 1024 bits. Landwehr. (1994). K. The six most often encountered during a one-month study done in the UK and reported for their percentage occurrences are: Forms 19% Parity Boot 16% NYB 9% AntiEXE 7% Sampo 5% Jack Ripper 5% Sources: Virus Bulletin. and Cheswick. (1993). Journal of Computer Information Systems. July. The globalization of computer crime and information security. L. p. R.R. 32(9). 61 74. IEEE Communications Magazine. A. 17(2). and Hsu. (1991). 1 12. D. (1994). (1990). 203 206. Hunting down the hackers. (1995). Computers and Security. 19(1). and Computerworld. IX(1). (1994). O’Mahoney. W. and Sandhu. Arjen Lenstra. R.E. and Farrell. 27(3). 32(9). Oct. (1994). Security across the LAN. 12 17. B. Oxfordshire. McDermott.Security for telecommunication jewels were sold and the bank was the first to make money out of a fraud committed on it. Computers and Security. Sherizan. 47 50. Access control: principles and practice. J. Cyberpunk: Outlaws and Hackers of the Computer Frontier. Molva. P. UK. Byte. S. H. Network firewalls. Computer viruses: symptoms.M. Introduction (to special section on securing cyberspace). Sandhu. September. A taxonomy of computer program security flaws. Bird. Gassman. Ganesan. and Nash. Byte. and security. Source: Byte. (1994). Security strategies for the 1990’s: security as an enabling technology. (1994). Spring 19 22. Computer viruses: can they be prevented. Supplement 12. R. XXXIII(3). (1992). Scientific American. 18(6).P. Cobb. 90 101. Stealth virus attacks. W. P. 1 15. The Canadian Journal of Information Sciences. E. VII(2). Enhancing the security of network systems. Bull. 154. P. p. (1993). 13 20.. (1991).S. R. 13(2). Siemens Review. S. 40 48. Symonds. 15. In 1977 they offered a $100 reward to anyone who could break their RSA-129 code (named after the 129 digits (429 bits) long code). 1995. S.

1993: p. ž Perform maintenance when needed and preventive maintenance to avoid or at least reduce breakdowns of system. ž Monitor system to identify potential fault conditions and ‘fix’ faults in real-time with minimum loss of performance. MANs for metropolitan areas and WANs for wide area networks. However. and so to grow. . As PCs became more robust and user friendly.13 NETWORK MANAGEMENT Network bottlenecks could be choking your company’s power to operate efficiently. Function smoothly. ž Provide statistics (on system and its components) necessary for planning and control of system. As we begin connecting departments together. there was need for an organization of management of networks that achieved most if not all of the following objectives: ž ž ž Increase productivity of end-users. Facilitate cooperative work between end-users. defective devices. and incorrectly connected components resulting in 140 crashes of networks. databases and computing power that they needed. devices that flooded networks with junk signals and slowed systems performance. a network is a set of nodes connected by links and communication facilities that have both physical and logical components. loose connections. Achieving some of the above objectives is the function of management of networks and also called network management. to survive. the number of PCs on desktops of most corporations increased. They had to pool and share resources. ž Detect fraud activity but verify and account for selected legitimate activities. Stafford Beer Introduction In the 1980s. The solution to the connectivity problem was networks: LANs for local area networks. With trends towards decentralization and deregulation of the telecommunications industry in the US. ž Implement and enforce security zones around sensitive resources. ž Be able to effectively integrate within the corporate information system. The new world is characterized by the need to mange complexity. In this context. It is the subject of . Managing the network becomes a strategic issue for the whole company . 277). The increase in the number and use of PCs led to the complexity of networks. ž Control devices remotely if necessary as in cases of failures. You need to manage and control networks where zipping ‘across your network are thousands of packets of information.’ (Derfler. Losing even some of that data can cost your company millions of dollars. many corporations could no longer afford the desired peripherals. efficiently and effectively without loss of integrity and security of system. especially the prices of PCs. To restore order to this chaos. overloaded components. Many of these resources were dispersed and had to be connected. continuously. ž Be robust against errors and misuse. to export. the network becomes a mission critical resource. innovation increased and prices dropped. concurrent access. Chris Bidmead The old world was characterized by the need to manage things. ž Plan hardware/software configurations for growing needs of applications and information. There were bottlenecks in message flow. Information that’s vital to your organization. we witnessed the strong emergence of PCs (personal computers) as stand-alone computers on desktops. .

A filter allows you to sift through traffic for a selected set of parameters such as addresses. alarm or action taken. It can search for duplicate network addresses. Such information is not only useful in tracing the cause of the problem. Security management. . provides more information and for the entire network. filtering. One type of analyser is the packet analyser which identifies the device that is clogging the network. and devices so that they can be initiated. Some analysers are portable ones and are dispatched to a trouble spot. Then there is the analyser. This may not be adequate in a continuous process production line especially with a high value added product in which case the analyser is dedicated to a function or process and resides in a probe attached to the LAN. and collect clues on how the network can be improved. Accounting management. but in helping maintenance and in trying to ensure that the problem does not occur again. In it. They could immobilize an organization and the problems must be quickly detected and corrected. Like the traffic monitor.Network management this chapter. there are other pieces of information necessary for fault management. identify the probable cause of failures. Fortunately. 5. We will examine how our networks can be safe and running. alerts and alarms generated. We conclude with an overview of the process of network management and look at future trends in networks and network management. with each problem that occurred. offering graphic representations of data and alerting errors to the operator) can be distributed between the embedded analyser and the manager software at the centre. providing frequent or continuous information to the central network management. the managing of the human element in networks management. In addition to the trouble ticket and analysing tools. In contrast to the poller. 2. there are other facilities for fault management such as the trouble ticket. Fault/problem management. In addition to the tools discussed above. we shall examine the functions of network management identified by ISO (International Standards Organization) as: 1. frame types and even devices by the manufacturer. It records the time. its tasks (of monitoring. isolate failing nodes. A traffic poller sends echo packets to a specific device to check if there is any fault in the transmission line. analysing. These problems could be a loss of performance. one or two protocols. closed down or restarted 141 In addition. traffic densities. operator. 3. we will examine the software necessary for network management. Performance management. may need control over the network. the development of networks and the acquisition of resources needed. Analysers can be specialized like the protocol analyser which can trigger a specific action when a preset threshold is exceeded. monitors and analysers. sometimes called the LAN analyser. as well as notices. impaired transmission or a systems crash. using a large number of predefined filters. The network administration staff in testing the system and reactivating it. If the analysers use a distributed client server architecture. Also recorded is the equipment involved and its vendor. one needs the hardware software platform handling high density traffic which may be around 140 different protocols blazing across the network at any one time. we have many aids and tools to help identify and trace a problem and to provide statistics for maintenance and planning. On the low end. There is a wide range of services offered by a network analyser. its links. we have the software analyser that will be adequate for a low density of traffic (around 50% of capacity). network performance. Such analysers can perform remote monitoring and control. The analyser for a network. trouble shooting between two network nodes. A traffic monitor is concerned with actively checking for faulty devices. But for more complex systems. Management of networks Fault/problem management Perhaps the most important and certainly the most urgent function of network management is to detect and fix problems on a network. Configuration management. These tools include pollers. These include aggregated and disaggregated statistics on errors counts. and with limited functionality. This information should be available in easy to digest format in addition to graphical map representation of the network to locate problem areas. the analyser is passive. a traffic monitor passively ‘listens in’ while displaying histograms of traffic patterns. date. place. 4. but provides more information about the packets.

We do that in business when we need to restrict access to certain buildings. There may also be eavesdropping. routers. for example. disks and other memory devices. This capability. For example. Another approach to security management in networks is to build a ‘firewall’ referred to briefly in the previous chapter. Hence there is an overlap in our discussion but that overlap could serve as a review. Also. other potential problems. monitoring and metering of variables relevant to operations. clients. and areas that need expansion or contraction. statistics on the use of licensed software can be very helpful in deciding how many copies of each licensed software to acquire rather than support clients with dedicated copies that may seldom or never be used. software utilities and application programs). the monitoring of a file system may start when it is 80% full. Security management involves the definition of the jurisdiction of network zones. can be quite costly to the organization. where it refers to the technique of preventing a fire from going beyond a line of defence. modulation. The variables include network (and its components’) availability (which is the mirror image of downtime of the system). some faults can be isolated all the way down to a particular segment of a network. and also help predict future trends for planning and budgeting of future systems. For example. This is a tempting occasion for penetration of the system. special problems of security that arise in telecommunications and networking because of the exposure of messages during transmission. multiplexers. modem or other device on the network. A more secure approach would be to page the person being visited who would then escort the visitor in the Performance management The management of performance involves measuring. and servers). However. for example. An illicit user taps the computer when a bona fide end-user is connected to the system. Such statistics when properly analysed can identify bottlenecks (current and potential). or active wiretapping involving alteration of data. switches. During operations. printer when the print queue is longer than five or ten minutes’ wait. utilization of various devices and resources: both hardware (like CPU. the electromagnetic interception of messages on communication lines. but while ‘thinking’ leaves the computer idle and unattended. such as piggybacking (the selective interception. or for a 142 . often remotely from the central network management console. security management may require enhancements to existing features and the addition of new features. communication cards. Metering and monitoring is not just ‘Big Brother’ snooping but provides data and statistics necessary for operations. Each hardware device is associated with specific network software and protocols. as well as traffic density by segments of the network. modems. The problem of detecting faults can become complex because many network systems (an organization may have more than one) not only have many different devices. Another type of infiltration is reading between the lines. licensed software can be monitored when. Many of the problems of security (and their solutions) are common to IT. buffers. or substitution of messages). which along with other unauthorized uses of computer time.Telecommunications and networks remotely from the central console. as well as specialized techniques to interpret the alarms and reports generated before other appropriate diagnostic and tests are performed. and software (like NOS. prevents the users in one department from using an optical scanner and printer in another department. Monitoring can be selective and activated by a set thresholds. There are. repeaters. Security management We have discussed security management in another chapter but that was largely from the view point of planning and design. as are the cases of controlling access to a database. gates. The term is taken from fire-fighting. 85% of its privilege has been exceeded. or even a device and peripheral. Likewise. invalid configurations and frontpanel tampering on various types of equipment can be detected. through the use of loop-backs and tests. an applications program. which are available to everyone or only to users in selected zones. maintenance and the planning of a network. but many of these devices may be from different vendors. One approach would be to demand an identification and logging of all those visiting the building. response time (the elapse time from query to output). bridges. They should also be able to divert traffic from failing lines and devices without the end-user being inconvenienced or even knowing about it. The principles of security management were discussed in the previous chapter. passive listening. however. Messages are vulnerable to wiretapping.

This automatic rerouting is very important for ‘mission critical’ applications where unnecessary delays and outages cannot be tolerated. Security management is also responsible for backing up the system in the event of a failure or a disaster on the network and to do so quickly and without loss of valuable data/knowledge. mentioned briefly earlier. the population of ‘unknown’ viruses is increasing (six were added each day in 1991) and it seems that a network system is never completely secure. Also. or just not use the systems as much or not use them at all. ignore the systems procedures (such as being careless with the passwords). Fortunately. Fortunately. Accounting management This is not accounting in the financial and auditing sense but an accounting of network assets. installing new software packages or their updates. A lot of network software is able to scan for known viruses. especially an end-user who has a computer at home and swaps floppies from home to the office. there are many protective procedures that should be adopted against viruses. then it is not only unnecessarily expensive to implement. Network management needs to be increasingly and continuously vigilant. These numbers are then used in collecting statistics on the operations of the devices. intrusions. and not leaving the computer on and unattended all night or during the lunch break. In security management. Reconfiguring may mean the addition. either through different links and connections or through different permutations or combinations of equipment devices. all these approaches to building firewalls are used depending on the preference of the network manager. which means that more than one router is often required. Configuration management All network systems need to be initialized and then reconfigured. or at least minimize risks of security infringement in the future. This means that careful logs must be kept of all failures. Access is controlled especially at the connections between networks such as a proprietary network and a public network. withdrawal or modification of an existing configuration. there are numerous virus detecting communications software products with different types of scanners (including signature scanners). However. A still more secure approach would be to not permit the visitor to enter. Routers needed for reconfiguration are protocol-specific. These include: not putting executable files on the server in directories where end-users can change them. then there are violations by intruders and there may be some who are waiting for the opportunity. restricting dial-in access. New technology. there is now network software that does the distribution automatically across all legitimate workstations and clients. Reconfiguring a system is also necessary when one needs to reroute networking traffic. In addition. to configure a device means assigning it a zone number and network number for routing purposes. if the virus gets on a network the contamination can be great because the propagation can be fast and spread widely.Network management building at all times. In networking. Reconfiguration can also include the rerouting or bypassing of traffic from overloaded or failed (or failing) links or congested devices to others that have slack. If security is too tight. the programs loaded onto each client node automatically draws updates from the server. In practice. Empirical data shows that a virus often infects a corporate database through a corporate enduser swapping floppies. In the early days of computing. the fileinfecting virus and the memory resident virus. is delivering multiprotocol routers which are capable of routing several protocols simultaneously. This is not an issue new to IT and arises even with stand-alone systems that do not use networks. Another problem is with viruses. the original configuration is restored. a network manger went around each office cubicle with a computer. Another responsibility of security management is to record all information that may be useful in the future. 143 . If security is lax. Another function of configuration management could be the reconfiguration of the application programs and their updates. memory resident activity monitors and immunizers. Other common viruses on networks are the boot sector virus. but also has a high psychological cost of alienating the end-user who may then either bypass the system. This saves a lot of time and energy. but to leave a message (or package) at the front desk. it is important not to overspecify or underspecify. When the quality improves. But. and qualifying each with identifiers that will trace each problem and help avoid them. and unauthorized access. however.

Such a choice is not unique to network management. Such systems will keep track of use under the licence so that violations are avoided or at least kept to a minimum. The accounting of each asset should be by physical location as well as by ownership so that each asset can be located quickly when needed (as in the case of a system crash) and for purposes of planning and operating the system. Some of these features are available in stand-alone packages.1. It is therefore sometimes necessary to enforce the limitations of use and is often done by warnings when thresholds are repeatedly violated. User management Besides hardware and software. Selected features of network management software are displayed in Table 13. Furthermore. Accounting of software is also important when the software is under licence. Every user of a PC has to face the decision of buying the best software packages for each function like word processing. Nowadays. However. These features match most of the features in the list of functions discussed above (though not in the same order) as being part of the functions of network management. buying an integrated packaged suite that has all the desired features plus much more. but it does 144 avoid some of the problems of making the different packages compatible with the operating system available and with each other. database management and even networking. It is therefore up to the network management personnel to select one or more software packages or to acquire a suite of integrated programs which may be more expensive but more comprehensive. This tracing of licences is important for the purpose of paying for the licence as well as for ensuring that no legal requirements are violated. The suite choice is expensive and not optimal for each function. the format for each package is consistent and one does not have to learn a different set of commands and icons for each package. the functions needed by any one network will be unique. much was done manually. This is the equivalent problem faced by network managers: acquire a suite that does all or most of what you want and pay more. spreadsheets. Accounting of assets is also important for financial accounting where the use of resources may be charged back to the user for payment toward maintenance and upgrading. but do not have the hassle of packages being incompatible and having to learn more instructions and formats. Some features are ‘bundled’ in other packages.1 Selected features of network management software Diagnostic analysers Server monitoring and reporting CPU utilization memory utilization log-on and security provides configuration information Network traffic monitoring Application metering enforces blocking when licence is exceeded queues users when licence is exceeded Application distribution Notification and alerting Virus protection server client Software distribution Hardware and software inventory allows setup of licence limits Reports predefined user-defined filtered Automation and scheduling of tasks client automation server automation remote control of client PCs Software support Integration User interface Database Free technical support Software for network management We have discussed the many functions of network management. there is another important element in every network system: the end-user. almost all such work is done by network software. But how are they implemented? In the early days. Some of them are professionals . Table 13. alternatively. And some suite packages have most of the features.Telecommunications and networks both hardware and software. or.

especially the interfaces. if there is a system crash. This is the advantage of the open architecture (where architecture is the style of construction of structure including layers in the construction) and open systems. The activities of the SDLC for a network are similar to other projects in IT: feasibility study. user’s requirements specification. or considers the existing LANs to be inadequate and needing to be replaced? Then we need a project which must be developed. implementation and the system made ready for operations. Take for instance an office with a large volume of interdepartmental mail that was traditionally delivered manually by a messenger. design.Network management with considerable knowledge of computers and telecommunications. there is a profiler program that collects information provided by the end-user. However. not being tied to one vendor has the great advantage of being able to select the best (or cheapest) component and ‘plug and play’. The advantage of having one vendor is mostly in the short term since this will avoid the hassle when something goes wrong and one vendor blames another. During operations. For example. like the type of LAN that is best for the needs at hand. causing a drop in productivity. End-users need to be trained not only in accessing and navigating networks. then the system is redeveloped. but the immediate problem is having to learn about telecommunications and networks and having to change the lifestyle at the work place. the system should acknowledge the problem and estimate the time it would take to come up again. made secure. ISDN. some psychology and a lot of patience. Introducing personnel to networking involves some of the same problems of introducing lay persons to computers. tolerance of common human errors. These activities and their iteration and recycling is shown in Figure about one that has a network but wants another LAN. Another important consideration in the design of a network is to decide whether the entire system is to be supported by one vendor or by many vendors and whether to make it an open system. If a modification is required and it is minor. End-users should also be trained on computer programs for network management. The choice is between LAN. maintain and manage. on trouble shooting and recovering from a crashed system. there are evaluations. For example. breakdown or slow-down in the network. There are also choices within these types of network. but also on the policies and procedures of handling data files. the origination of the network (should it be centralized or be a client server system?) should be considered. use of networks can be a cultural shock. then the system is maintained. They have to be trained and educated. and one that keeps the end-user fully informed. There are some differences in content with a typical IT project and these differences will now be examined. The first decision to be made is what type of network is needed. were end-user friendly. public packet switching network and a public switched telephone network. Systems Development Life Cycle. how about an organization without a network? Or. Many end-users are tolerant and understanding of problems provided they know what is happening and what to expect.1. on protocols (rules and procedures that must be followed in transmission). The change has to be managed with great care. then programs and updates due to the end-user will not be forthcoming. nor are they obvious. In contrast. The advantages do not appear instantly. The methodology appropriate for a complex project like a network is the SDLC. For example. a closed architecture 145 . where there is interoperability between components available from different vendors (preferably internationally). accounted for and maintained. This requires a development methodology. Now this is to be replaced by email. train on. training on the applications specific to a LAN should be done before the LAN is installed or else the ‘shock’ resulting from an unfamiliarity with the new application can be demoralizing. What does this mean? An example would be a system with a fast response time. There are. and on the functions and limitations of routers (facility that selects and provides a path for a message). If not completed correctly by the end-user. If the modification required is deemed to be major. many who have little or no knowledge of computers or of telecommunications. It would help if the network system. on the nature of network maintenance and network security. With only one vendor the system is well integrated and easy to install. But. The timing of the training is also important. Also. Development of networks Thus far we have assumed the existence of a network which must be reconfigured. For some. however. proprietary network.

is the CMIP (Common Management Information Protocol). reduce response time. though. MAN and the WAN). links to carriers. there are several network management platforms. The implementation of a network will most likely involve acquiring (‘buy’ decision) some network software and the rest is developed inhouse (the ‘make’ decision). XNA by Xerox and those developed by governments like the DDN in the US). At this design stage. Such software is also concerned with protocols such as SNMP and CMIP. OverView is actually a family of over 130 solutions from over 80 vendors. switches.e. that are based on standards including SNMP and CMIP. Europeans also supported standards developed by CCITT (Consultative Committee on International Telegraphy and Telephone) and ECMA (European Computer Manufacturers Association).1 Development of a network is proprietary technology developed by firms in the telecommunications and computing industry like SNA by IBM. security features. the chances were that you would have selected SNA. Network managers had to decide what products to add so that these would cause the least disruption when a standard was ultimately adopted. you would have preferred OSI. The important design decision of selecting an architecture often depended on where you were and what equipment you had. reduce costs and add to the availability of standard network products. Another important set of decisions concerns software. However. and even surge protection. Most of the hardware needed will be acquired and only seldom made in-house. i. . transmission mode. the CMIP and the SNMP protocols will converge. there was strong pressure to adopt international models such as the OSI model. The CMIP. the LAN. It is likely. then you would have selected TCP/IP. we are not concerned with software required for management of networks which was discussed earlier. If you had a UNIX operating system. The OSI is developed by the ISO (International Standards Organization) and was strongly supported by many countries in Europe. maintenance and administration User Requirement Specifications OPERATIONS Design Implementation No Evaluataion Yes OK? Feasibility Study No Minor Modification ? Yes Maintenance Figure 13. though. metropolitan and wide area networks. The SNMP is the Simple 146 Network Management Protocol which defines systems/network management standards for primarily the TCP/IP-based networks. such as OpenView. education. In competition. has not yet caught on.Telecommunications and networks Organization of personnel. but software required for operating a network. however. Other important design decisions include the selection of bridges and gateways. that in the late 1990s. if you were in Europe. back-up and recovery procedures. The OSI model would add to flexibility. But all over the world. there are well tried and true procedures of resource acquisition in IT that are applicable to network development. currently the only internationally ratified network management protocol standard for the OSI. The CMIP was intended to support business at different levels (local. DecNet by Digital Corporation. Meanwhile. If you are in the US and have IBM equipment. In all cases. The TCP/IP (Transactional Control Program/Internet Protocol) is a system designed for the US Department of Defense and is a commonly used communications standard. including the US.

This is an important and difficult decision. How soon can I have it? Of the above three requests. there are evaluations. The console operator should be able to see the network displayed (preferably in colour) at any level of aggregation. That depends on the network management systems program being run. The status of each crucial physical and logical component can be traced (each physical and logical unit may well be identified by its parameters such as name. physical location. especially if there are hundreds if not thousands of nodes connected to the network and when the operations. may depend on the network(s) for the enterprise. identifying the current and potential bottlenecks which may be flashing. I would like a LAN connection for my assistant who now is located in Room 39 of Building 131. it is a major problem of modification and may well require redevelopment.). Otherwise. the first is clearly a problem with maintenance and can most likely be covered by the maintenance budget and existing maintenance personnel. I think that I need a token ring network for my department. software and personnel.Network management During operations. The third request requires a major modification entailing new project development and new funding. but the terms of defining a minor or major modification are different. . The second problem is minor if Building 131 is wired for a LAN and is connected to the corporate LAN. Hardware and software Hardware and software are needed to provide information necessary for network management. and indeed the survival of the enterprise. Some systems may even have a simulation program that would give answers necessary for smoothing the load and increasing the efficiency of the system. All monitoring information and even statistics should be available that are either menu driven or command driven. etc. is a set of guidelines for policy and procedures relating to maintenance of networks so that the maintenance process is clearly stated and known to all end-users. What is the information that is needed? Much of it is needed for operations at the console on an online real-time mode. three requests for modification of a network: 1. Here the decision needs to be made as to whether a modification needed is a minor one for maintenance. Such maintenance decisions are not unfamiliar to IT personnel. My LAN connection worked fine for a year and now is down. The problem facing network management is to distinguish as clearly and quickly as possible which is a minor and which is a major modification. What-if questions asked may include: What if I added a workstation (or ports) at point 119? How would this affect the service time and length of queue? What if I added a router on segment 5? How will the performance be affected by the addition of specific hardware or software? What is the maximum distance travelled by end-users if I were to add a workstation at point 146? (The answer to this question will require 147 Resources for network management From the description of the nature and functions of network management it becomes clear that it is an important responsibility. state. 3. The console operator should be able to get answers to questions like: ž What are the network loads on different sectors and at different times? ž What types of errors are occurring and where? ž What types of conflicts (like concurrency) are occurring and where? ž What channels are loaded or near loaded with what load factors? ž What are the waiting-line queues and to what extent are they exceeding or approaching the set thresholds? ž Where is security weak or possibly being violated? ž What printer and print driver is location 119 using? (Are they compatible?) ž What is the printer and print-queue status of B562? Answers to some of the above questions and more may come without a query. then. The current network is inadequate for my needs. It is also a difficult one. This request needs further investigation and is placed under advisement. These are discussed below. for example. What network management needs. The resources needed are hardware. or a major one for redevelopment. Can you please help? 2. Consider.

Some systems have alarm filtering. there is need for adapting the organization to changing conditions. The open standard will not only provide flexibility in operations. As a result. be knowledgeable of accounting and budgeting practices. Unfortunately. additions must be balanced by subtractions. but withdrawing or not assigning high security levels of new software can cause ill feelings.2. but not to the supervisor. Some of these incoming messages are recorded and appear as reports for later analysis. To perform the functions of network management. there may be just one person (carrying a screwdriver) responsible for a network. Some systems can even predict network faults based on historical data. The personnel involved are of course different and must be 148 . These functions could result in additions and subtractions of nodes on the network. fall on the network administrator who. however. For example. the staff have to be organized. improve response times and reduce costs.) The answers to some of the simulation questions may be long term solutions that are important to planning. i. The staff can be organized either centrally or in a distributed fashion. We discussed organization in an earlier chapter but the emphasis then was on starting a teleprocessing department. This may not be a zero-sum game. each person may represent one function or multiple functions. but will decrease delays. We already have some intelligent components like the intelligent router. developing and operating a network. The functions of network management have been mentioned as being reconfiguration.Telecommunications and networks a database of floor plans. One such configuration. and the personalities involved both in IT and in corporate management. along with the other decisions needed for planning. one person may do other jobs as the need arises. grouped for functional efficiency and effectiveness. is for a network of a medium sized company. a network management centre for a large and complex system would have multiple consoles each manned by a trained operator and performing one or more functions. in complex networks the number of personnel involved may be in the tens. One desired feature of a network system on the wish list of many network managers is that network management systems programs be intelligent (by using AI techniques for making inferences). In the case of a small organization. shown in Figure 13. which could route a message between Paris and Frankfurt through New York. Also. In it. monitoring and maintenance. Many of these problems of network management will come close to resolution as we approach a truly open system (some systems use a buffer layer to approach an open standard architecture). the organization culture of the enterprise. there is a need for staff headed by a network manager. partly because different components of the system are manufactured by a proliferation of vendors. The organization will also depend on the size and complexity of the network organization. have a working knowledge of hardware. Such decisions. there is no system that is fully integrated that provides an end-to-end management. also called a network administrator. Security levels can be a status symbol. site plans. However. Some consoles have the ability to send and receive messages with the receiving being prioritized. This can be a difficult political decision in cases of a zero-sum game. Which one goes and which one stays? A somewhat similar problem arises in assigning software and security levels. The principles involved in the selection of a central or distributed organizational configuration is no different from those relevant for a distributed IT organization. and equipment inventory by location. but such capability is often useful in problem-solving and decision-making for a network. and also be a politician. wiring closets. in addition to being a technician. as the volume of processing increases and as it becomes more diverse and complex. a software or hardware engineer assigned to maintenance Personnel To perform the functions needed in network management and to manage its resources. Some perform functions of monitoring and reconfiguration automatically. and if a higher level of security (or new software) is given to a subordinate on a need basis. must be comfortable with software (especially operating systems software). the librarian or the security officer may be part time jobs assigned to a person most inclined and trained for the task or the assignments made in order to ‘smoothe’ and distribute the load. What we do not have is an intelligent hub and intelligent network management systems. serious problems of unhappiness and loss of face can result.e. For example.

Network personnel. there is a lot of swapping of data. Networking is not just networking of telecommunications and computer equipment. not only because the enterprise environment changes. but also with the planning of networks. Typically. Network management personnel have a virtual job that changes all the time. Their field is ever changing with changes that are technological with organizational and social implications. but also for the organization (providing back-up and knowledge necessary for integration). should it be wired for networks? If networking is needed after the building is built. yet the principles involved apply to other IT functions such as those discussed earlier. then wiring will be much more expensive. For example. This is important. Training is another important function of network administration. but for the nation and indeed all of us. this should be transparent to the end-user. but it can facilitate group work and impact on decision-making and problem-solving. for example. It not only brings people together through e-mail and bulletin boards. including the network manager and network planners. Network management must keep well abreast of these developments and new networking products as they occur and in some cases before they occur. This may be said of many an IT professional. Likewise. if the organization is to build a new building. Telecommunications is one of the sectors of computer technology that is changing the fastest and this can have profound implications. but more so for telecommunications personnel because their technology is ever changing and changing very fast. it is the networking of people. putting cable conduits underground in order to connect buildings for networking is much cheaper and easier if planned before the buildings are built. Also important is the training and upgrading of network management staff. Successful networking will require changes. They often get nervous with obvious changes and worry how they will be affected. Network management must not only be concerned with the operations of networks.2 Organization for network planning may well help out in the installation and implementation of a system and may even contribute to a feasibility study. This rotation of work is good for both the personnel involved (providing variety of work). The satisfied 149 . This may include facilities planning. including the network administrator. but also because their technology changes. changes required by corporate management and end-users as well as those required by maintenance (preventive or otherwise). from the client to the server and vice versa. networking should be planned because such planning ahead is not only cheaper but much less disruptive for the organization. Some of the changes in telecommunications and network technology will now be discussed. such planning is not possible. not just for the enterprise. but when possible. this means the training of the end-user and education in telecommunications and networking for corporate management and other corporate personnel. In all cases. Sometimes. Also. must be crosstrained on the central network console.Network management NETWORK ADMINISTRATOR Planning and Budgeting • Design and Implementation • Acquisition Hardware Specialists Operations and Maintenance Planners Economists Technology Watchers Financial Specialist Software Engineers Network Specialists Hardware Engineers Network Specialists Administration • Personnel • Security • Standards • Education • Change Management Administrators Specialists Figure 13.

) PC Smart Terminal Workstation PC connected (Intra Organ. They can often start and stop any client and any server. These personnel may not use their power and commit computer crime but they still have access to enterprise and corporate information that they may misuse to affect decision-making and problemsolving by corporate management. connected to LANs. Alternatively. They can ‘listen’ into any message that comes in or goes out. Training and education should be reactive. One way to overcome this problem is to build trust among the end-users and corporate management so that such misuse could never happen. they should be given what they need and when they need it. as well as cellular and wire-less devices. not too much before being needed and certainly not too late. The customer and end-user may also feel that the provider can potentially reach further into their domain than they would like.Telecommunications and networks end-users are often ones that are not required to change what they do not initiate and want. there may be in-between configurations based on CPU hardware (mainstay) Telecommunication configuration LAN//MAN/WAN/Internet Client−server system Open system Multimedia Cellular Organizations Interconnected (Intra Organ. There is often tension between the provider and customer because the customer often feels that the provider is not giving all that is possible and something is being held back. Education and training. as well as communication between the end-users and the technical personnel. But. are important strategies to reduce the stress of systems change. A state-of-the-art configuration might consist of a multimedia open system with clients and servers using PCs. They have complete access to the corporate and enterprise information that enters the network. Selecting the best equipment configuration for networking is another important responsibility of network management. The provider has more control over the customer’s equipment than the customer would prefer. they do not always want to know about the bits and bytes of the system. The relationship between the provider of communication services and customers of services must be reactive as well as proactive. These personnel are what Boguslaw calls the ‘computer elite’. This situation arises with other IT personnel especially database personnel. Also. Some management find it prudent to bond all telecommunications personnel. workstations and smart terminals. There is sometimes a tension between endusers and corporate management on one side and network personnel on the other side.) Data/Telecommunications network Fibre Optics Mini Mainframe Standalone systems Telephone PBX Mainframe Figure 13.3 Evolution of networks 150 . The customer may feel that their equipment is not restored fast enough and long after the systems performance is degraded. The other extreme could be the stand-alone system around a mainframe with a telephone and PBX connection. telecommunications personnel have great power of another kind. but also proactive. MANs and WANs. Why? Partly because telecommunications and networking personnel have great power since TIME possession of information leads to power. When they need to learn about the system.

Source: George Black. The system already has over 400 users but is designed to serve up to 4000 end-users. The plan for the ultimate network is still being implemented. This is technologically possible. Each set of configuration has its own growth curve depending not only on the network personnel but also the end-users. For example. Figure 13. The standard services include e-mail. and access to a CD-ROM library containing archives of national newspapers and Hansard. Which Computer?. mainframes and minis using fibre optics and a data telecommunications network. In many large organizations.1: Networking in the British parliament Networking has made a wide range of information services instantly accessible to members of the British Parliament. bulletin boards. Summary and conclusions A summary of functions to be performed by network management are listed in Table 13. The going is slow because cabling through the twelve inch thick walls of the venerable building is a slow process. each serving different units of the organization. Case 13. another more advanced configuration curve is Case 13. would Members of Parliament stay in their offices rather than go to the floor? Here is a good example where network management is not a just technological matter but a political one also. Fault/problem management Troubleshoot. All-party networking. But is it politically desirable? If implemented.2.Network management Table 13. By the time the top part of one curve is reached. However. It has a FDDI (Fibre Distributed Data Interface) backbone and spans three buildings including the Palace of Westminister.2 Summary of functions of network management. word processing.3 shows the evolution in network configuration. take the implementation of videoing all the happenings and debates in parliament into the offices of the Members of Parliament. there is not one but more than one configuration. any cabling should allow for the future. Knowing on which curve you are and when to switch to another growth curve or even leap-frog a curve is an important decision that network mangers must make. Also. budgeting and maintenance Security management Control of access to network Define network security zones Detect and protect against viruses Erect and maintain ‘firewalls’ Backup and restore Report unauthorized use and misuse Configuration management Initialize network entities Reconfigure systems (add/subtract/modify) rerouting of traffic through links and devices Account management Manage assets of inventory hardware software (application programs) Track and enforce licences User management Management of change Training Make system end-user friendly found that is overlapping and reaches higher levels of performance and service. diagnose and predict potential faults outages Identify. classify. some of the future services planned do not face technological problems but purely political and social ones.2: Analyser at Honda auto plant The Honda automobile production plant in the US needs analysers for maintaining its large 151 . analyse and report faults/outages Initiate selected automatic ‘fixes’ and restorals Performance management Measure/meter/monitor Network availability Response times Downtime Hardware utilization Software utilization Traffic Predict trends for planning.

we can toggle a view of any LAN. 51 52. Supplement 13. analyses trends and recommends improvements) $17K $15K $ 5K $ 5k $ 5K Source: Computerworld. and Heile.’ Source: Daniel P. Network management. 12(2). 40(1). B. Jeffers. 109 118. F. such as changing thresholds and trap requests. (1994). at Honda adds: ‘Using the master console. 93 103. Boehm. 237 267. S. (1997). IEEE Spectrum. or perform diagnoses . Managing networks. Telecommunications. F. (1988). network management ‘can make changes to the settings on the remote units. Henderson. And Ullmann. or all the LANs. Personal Computing.’ The site has a LANvista system that comprises a master console with four remote token ring ‘slave’ probes. ‘so we can get immediate alerts of problems and already have the diagnostic equipment in place. Communications of the ACM. PC Magazine. 8 15. G. 27 34. Nov. 55 58. 277 300.S. 53 56. ATM: will it live up to user expectation. C. LAN. Which Computer?. 29(6). Rigney. 538 551.A. 12(7). Macworld. Broadhead. An eye into the LAN. (1992). P. C. Telecommunications. Network quality assurance: a checklist. (1991). Telecommunications. Steinke. and Pervier. Fundamentals of network design. Since it has a high value added product it cannot afford delays in its monitoring efforts and in detecting failures in its token ring network.J. Wiley. 57 60. LAN tamers. Corrigan. 12(1). Tutorial on SNMP. Chapin. (1993). Marx. Managing network stations. L. PC Magazine. 4(1). S. 46 51. Digital access devices: criteria for evaluating management options. (1991). (1997). N. LeFavi. (1997). (1994). Nachenberg. can monitor and control multiple screens from a single console.Telecommunications and networks network. we may already have data on problems which is important if something has crashed to the point where we can’t recreate the event’. (1994). 27(6). 29(7). Management wanted probes connected to its critical LANs all the time. p. 27.1: Prices of LAN management software in 1995 Problem management (including the providing of monthly reports) Remote monitoring Administration (manages user IDs/passwords and resources) Backup and recovery Performance tuning (tracks baseline performance. C. 12(1). Computer virus-antivirus coevolution. ‘Troubleshooting remote LANs’.L. 14(20). 152 159. Strehlo. Derfler. 13 18. G. Schneier. 29(4). 15(10). February. New telecommunication technologies require new manners. A 10 point prescription for LAN management. 21 22.T. Telecommunications. International Journal of Computer Applications in Technology. Applied Cryptography. Dern (1992). (1995). (1993). Bibliography Antaya. Because probes are continually monitoring. (1992). from the master console. Network management. R. Muller. A. P.J. Telecommunications Policy. . D. (1992). Jr. (1995). ‘The state of the Internet. Kosiur. Because of the client server architecture used. Datamation. D. Information Systems Management. . 1995. Briscoe.54.B. to view any network activity. 11 125. 28(1). S. (1995). 38(4). 9(4). 152 . LAN. Integrated network management. 18(7). W. 25 30.

i. and software that facilitates distributed processing of applications programs. This is the sort of promise that parallel processing hardware offers for AI. at a finer level of detail. there are both coarse. if the AI system has to search many options in order to find a ‘good enough’ one.e. each one processing a different alternative decision.and finegrain parallelism to be found in AI problems. and two. However. Patrica Introduction We have discussed some resources needed for teleprocessing such as the media for transmission. hardware that can use more than one processing unit on a given problem at the same time (in other words. Thomas J. A coarse-grained parallel implementation of such a problem may be able to use a small number of processors all working in parallel. The machines are simply tools which man has devised to help him do a better job. At the so-called coarse-grain level a problem is composed of a small number of significant subproblems that can be processed in parallel. Some processors will come with their own operating systems for telecommunications.e. we have PCs and workstations. processors working in parallel). So. PCs and even minis were no longer adequate and we needed mainframes and file servers. It is these computing processors that we will examine in this chapter. At the server end. Such software we will examine in this chapter. One is that parallel processing will be the faster way to perform a large volume of computations simultaneously. But in addition we need software at the client-end. at the top level of the management decisionmaking task. There are two good reasons for discussing parallel processing in a book on telecommunications.14 RESOURCES FOR TELEPROCESSING This is a Man’s Age. Such processing along with data and text is the multimedia processing that will be important in the future mix of teleprocessing traffic. We also recognized that the traffic is rapidly changing from data to text and now to multimedia. network management resources. To begin with. Thus. and telecommunications personnel. the search can be executed in parallel. the devices for connectivity. and there are a number of very different ways to exploit parallelism in AI programming. referred to a comms software. i. At the client end. we shall not discuss applications software because that is independent of telecommunications and beyond the scope of this book. referred to as middleware. Watson (of IBM fame) I don’t think the mind was made to do logical operations all day long. But even this processing capacity may not be adequate and soon we will need parallel processing and perhaps even neural computing. because parallel processing is very appropriate for pattern recognition which is the procedure for voice and image processing and other applications of AI (Artificial Intelligence). there will be a number of quite distinct alternative possible decisions that could be made. then perhaps it can search more than one option at once. Parallel processing One solution to the demand by AI systems for raw computational power is to move to parallel architectures. As the complexity of the traffic increases and as the volume of traffic increases we need more capacity for the processing of all this traffic. But. At some point each independent processor must then report its findings as to the worth of 153 . We start with parallel processing. there are several rather different styles of parallel architectures. software in the connectivity devices.

) to a central processor. 1987) offer the programmer a large number of quite primitive processors. Hence the difficulty in designing systems to exploit such coarse-grained parallelism.g. Such parallel hardware is so important for efficient image processing (and pattern recognition in general) that there is a whole range of different hardware products designed for just these sorts of applications. these massive parallel machines are easier to use than the previous ones we have discussed. It is quite conceivable that an operating system could exercise this sort of judgement on a program that it is running. At the moment. Furthermore. the early stages of this visual processing e. of course. in general. this centralization of control was only necessary at the end of the task (and presumably at the beginning in order to decide how and what alternative decisions to explore initially). Notice that there is always a need for some processor in the system to assume exclusive control from time to time. These machines (such as the Connection Machine. its cost and side-effects. It commonly occurs in vision systems and also in many different types of so-called neural-network models the new connectionism. which will then choose the best decision to make. And finally. In the case of our simple example. this is largely an unfulfilled potential. To construct parallel programs appears to be much more difficult. involves large numbers of parallel processes (also 154 known as fine-grained parallelism a term that emphasizes the elemental nature of the parallel processes). parallelism. In fact. in this class of parallel system’s work. which made a small number of powerful processors available for the programmer’s use. So. it is often hard to predict. These are shown in Figure 14. and the computer system would. the hardware is well in advance of ability to use it effectively. as the name suggests. conventional programs) is in itself a very demanding task.1. exactly when and what will need to be communicated between the processors working in parallel. there is a continuing need for this sort of processor intercommunication throughout the process of complex problem-solving. Coarse-grained parallelism. So we can confidently predict the coming importance of this type of parallel computer in business applications of the future. and so we have an application of massive. etc. Each quite simple operation can then be performed independently of most of the other pixels in the visual image. although there are many connectionist models . see Waltz. And this is because the primitive processors either do not need to communicate with each other while executing. with its relatively small number of parallel processors. which.Telecommunications and networks the specific decision it processed (this might be in terms of how well it solves the original problem. The construction of accurate and efficient simple sequential programs (i. is sometimes known as mere parallelism to contrast with massive parallelism. and. The new AI subfield of connectionism is another potentially major application of massively parallel computers. although there are a number of system-development projects working towards this sort of goal. as opposed to the previous category. but simple. Then the programmer’s task would reduce to the conventional one of providing a correct sequential algorithm (with perhaps some indications of parallel possibilities). in advance. But it is very difficult to structure programs so that they can effectively exploit the coarse-grained parallelism inherent in a given problem. But. When does this type of parallelism arise? It actually occurs in a number of somewhat different ways in AI problems. A first point to clarify concerns the possibility of using such massive parallelism when we have just explained the great difficulties involved with the effective use of only a few parallel processors. parallelize the algorithm for more efficient computation. because. the operation to be performed on each pixel typically only involves the few nearest-neighbour pixels. It is relatively easy to construct computer hardware composed of a number of processors that are both capable of working independently in parallel and of intercommunication. the noise removal and edge finding often involves a very simple computation that has to be repeated on each pixel in the image. Computer hardware is available for directly supporting this fine-grained parallelism. In addition. You should know that computer processing of visual input amounts to the processing of thousands of simple data items known as pixels. The answer is.e. such systems are still largely a future event. according to its sophistication. for the computer system itself to detect and exploit the parallelism inherent in a problem as and when it occurs. or because the necessary intercommunication is simple and predictable in advance.

These include: operating systems for networks. and then there is middleware that that is the enabling layer of application program interface software that is in the ‘middle’ of the network and the application. is easy to see but not very productive.1 Sequential. software on devices especially smart and intelligent devices. Network operating systems Just as an operating system for a PC enables the user to access peripherals and manipulate data 155 . The unifying theme in this AI subfield is that the main computational effort is distributed over a large number of primitive processors. Nevertheless. The analogy with brain structure and function. sequential computers we expect this to change. neural-network models seem to offer problem-solving potential that is. there are no deep similarities between human brains and so-called neural computers or connectionist models. in some significant ways. schemes that are being explored under the banner of connectionism or neural computing. We shall discuss each below. The neural computers themselves have proved to be powerful pattern-recognition devices. most of them are being implemented on conventional. Software for telecommunications There are many types of software (besides applications software) needed for telecommunications and networks. the analogy is only superficial. parallel and massively parallel processing being built.Resources for teleprocessing One step at a time: traditional sequential architecture Several processors operating simultaneously and in parallel (mere parallelism. very different. networked together and operating in parallel. software that facilitates linking to say a LAN (MAN or WAN) called telecommunications or comms software. with its network of neurons operating in parallel and communicating via on/off pulses. superior to that obtainable from more conventional implementations of AI problems. Currently. coarse-grained parallelism) A network of processors interwoven in complex and flexible ways in massively parallel systems (fine-grained parallelism) Figure 14. There are many. called the network operating systems (NOS).

a PC would use the ASCII code. 156 A combination of features will perform different functions. a seven-bit code. Software for linking devices Most software for linking devices performs a specific function. ž X-ON/X-OFF support which allows flow control for accurate uploads and downloads. The comms program must recognize differences if any and make the necessary conversion for compatibility. there’s probably a comms package designed to meet it. ž Dialling a phone number without assistance. 210). This guarantees proper transmission and arrival. like a parity bit. Connection to a BBS (Bulletin Board System). There are some . word processing. modern computers include software with integrated applications or at least integrated comms programs and modems. In fact. ž Recall and perform a logon sequences of instructions for a remote computer system. Internet e-mail. like say a router. E-mail. parity. Another code used in a comms program is the control code. 1995: p. Some of the many functions include: ž ž ž ž ž ž ž ž ž Dial-up mail. Comms programs do meet most network problems. ž Compress data to reduce time and charges for transmission. There are. The author bought a PC in 1993 that had a comms program with its inbuilt modem (including a fax modem). memory) and rely on a remote host computer for all the processing tasks and the terminal then behaves like a computer to the end-user. and not have to reset every time. Comms software We have implied the use of comms software many times in earlier discussions.’ (Derfler. For example. A wide range of communication speeds. Break signal. In contrast. and data management. For example. Connection to the Internet. The comms software enables the linking of a computer to a LAN to perform on-line computing remotely. other applications that are telecommunications-independent. But the best bet for both the personal and the professional user is still a general-purpose package. which allows the sending of a signal instead of a control-key and letter combination to interrupt a mainframe computer. also called autoanswering. Network access. Set user-set defaults specifying speed. It has a combination of features that may include the following: ž Allow for a variety of transmission codes. However. also known as autodial. These include spreadsheets. ž Answering an incoming phone call at an operatorless computer system. Host links. a mainframe may use a six-bit EBCIDIC code. ž ž ž ž ž ž Encrypt and decrypt data for security purposes. The comms program may need to be integrated with these applications in addition to telecommunications applications. ž A second level of communication will be having the ability to send and receive data. however. need to perform the function of routing messages optimally if possible. Remote file access. Check file names on a disk without leaving the comms program. However. ž A low level of communication may be emulation as in say a terminal emulation where a terminal will bypass the local computer’s resources (like CPU. It controls access and use of a network and ensuring correct and hopefully efficient use of these resources. likewise a network operating system enables a PC connected to a network to use peripherals. Well how does a modem have these capabilities? Through comms software. in the discussion of modems we asserted that modems have the capability of error checking and compression. if ‘you have a specialized communications need. Changing settings while still connected to another system.Telecommunications and networks off a disk drive. Sending a file from a local computer (or client) to a remote host computer is (or server) uploading. On-line services We shall discuss the last five applications in later chapters when we discuss telecommunications applications. Receiving a file that has been transmitted by a host is downloading. some devices. that is. etc.

error recovery and transaction management. which enable the access by calling desired procedures located in different places on the network. network architectures and protocols.2 Middleware in the layered architecture 157 . data is scattered all across various incompatible networks. Middleware has file-transfer capabilities and can hook into email software for development of mail-enabled applications as well as support object-oriented programming techniques. which gives the programmer a single API that can be used to access different databases. Thus it needs to be able to make inferences given decision rules.Resources for teleprocessing operations research models that compute optimal solutions. This makes the device quite valuable because it no longer is ‘dumb’ in the sense that it must do exactly what it is programmed to do but in addition it is able to adapt to changing conditions. There are many types of middleware: Database middleware. The middleware is the interface between the sessions layer and the application programs. Also. applications program. When an RPC is invoked. MANs or WANs with varying protocol stacks. One can look at middleware in the context of the seven-layer ISO network architecture as shown in Figure 14. Middleware functions will vary with the vendor. which enables access to services offered by any object anywhere on the network. which means that the middleware will support the high level languages SQL found N Middleware software Middleware is the software enabling layer that supports multiple communications platforms and protocols. Object Request Brokers (ORB). Middleware is an application program that has APIs (application program interfaces) that enable working with separate programs that interoperate even when they are running on different LANs. a program executes a procedure call very much like executing a subroutine in our standard 3GL procedure language like Fortran or COBOL. The call is received by the remote server and the results are returned to the sender. computers. The essential services offered will include distributed computing with security. operating systems. location transparency. The APIs enhance the sessions layer functionality without reducing its functional complexities but makes it easier for the programmer at the applications end. The objects could be programs or they could be resources. RPCs are not very popular in a client server environment because procedures are not the best way that processes communicate in advanced operating systems software. In many computing enterprises. balancing of load. but in the sense that it has intelligence in the sense of AI. Structured Query Language (SQL) Orientation. devices could be smart not in the sense that it has a computer in it and can compute as a stand-alone device. Remote Procedure Calls (RPC). each corresponding to a separate Applications 1 2 MIDDLEWARE Sessions Layer Transport Layer Network Layer Data Link Layer Transmission media Physical Layer Figure 14. Here the network traffic can be seen to split into different sessions.2.

micro. i.1 Summary of Middleware characteristics APPLICATION ENABLING SERVICES Transaction workflow E-mail EDI (Electronic Data Interchange) DISTRIBUTED SYSTEMS SERVICES DISTRIBUTED SERVICES Location Time Security COMMUNICATION SERVICES Database access RPC (Remote Procedure Call) OBR (Object Broker Request) SQL (Structured Query Language) Message middleware Process-to-process messaging Message queuing Transactional message queuing Persistent message queuing Non-persistent message queuing cross-platform file servers and provides interoperability to total enterprise applications and network services (even something simple like printing). . the database middleware and the RPCs are synchronous and may stay idle until it gets a clearance and permission to transfer. 1992: pp. the effect is that of a distributed operating system not just as simple interprocess communications API . . . and mainframe over just about every protocol. Also: . In contrast. protocols. When middleware is at its best. After this call is made. ORBs or other software that needs to work in more than one place simultaneously. . that is. or message queuing. 66). survive failures.1. persistent message queuing. Message middleware is inherently asynchronous and ensures that a process is never blocked and processes the messages concurrently. we can say that middleware offers a high-level interface to 158 . Table 14. Different characteristics of types of messaging is summarized in Table 14. which do not survive across any failures. the goal is to find the API that has just the right level of abstraction and functionality for the job at hand . . Currently. ‘Data point’. the APIs enhance sessions layer functionality while reducing its complexity . For message middleware. . Message middleware comes in two types: processes messaging. processors will need to be powerful and versatile. but between them they address all conceivable mini. (King. mainframes are used as servers Summary on middleware Summarizing middleware. specify the sending and receiving data point after the command ‘SEND’. initialize the necessary protocols and ensure any recovery handling that may be necessary. which survive some failures and do not lose the message. the message-based middleware uses literally hundreds or thousands of lines of code to set up a session with the remote destination node.e. . 64. . which requires both the sending and receiving processes to be available at the same time. and non-message persistent queuing. . which enables programs to communicate via messages which is important in distributed computing environments. ‘Data destination’. Message middleware. a programmer who wants to communicate between programs such as send data from program A to program B will send a command like SEND. 63. which does not require simultaneous processes and can stack them in a queue. the basic chore of middleware is to move data . This is a great savings to the applications programmers and makes the process so much easier and user friendly. Message queuing implies a queue manager. . and programming languages. No single vendor of message-delivery APIs support all platforms. Thus the messaging middleware is more efficient than the database middleware and the RPCs. The message middleware can vary with queuing characteristics: transactional queuing. a software program that is responsible for supplying the message queuing services used by applications including the ability to store messages in a queue for later delivery. Summary and conclusions As teleprocessing grows in volume and complexity.Telecommunications and networks in many DBMSs (Database Management Systems) and other dialogue languages systems. A message middleware could establish a communications sublayer that could support other middleware such as the RPCs.

EEI took the existing Statistical Analysis System (SAS) and its COBOL applications 159 . programming languages and protocols involved. All these resources have been discussed in previous chapters. In hardware.2. The distributed environment with its many interfaces and services as well as development tools and Table 14. Modern teleprocessing has a wide variety of options for each of the resources needed. Whatever the choice made by the network manger. The controllers (IBM 3174/3274) and telecommunications hardware which was older than the 4381 proved to be worth more and collected a dew hundred dollars as salvage value.Resources for teleprocessing of teleprocessing services as well as for applications that need teleprocessing. In connectivity. 263). the processing needs will increase. In architecture and protocols.2 Summary of Services and Languages in a Distributed Environment SERVICES Transactional management Transactional queuing Messaging queuing Messaging middleware Protocol stacks Error checking Resource recovery Security DEVELOPMENT TOOLS 3GL (Third-Generation Procedural Languages) 4GL (Fourth-Generation High Level Dialogue Languages) SQL (Structured Query Language) CASE (Computer-Aided Systems Evaluations). access methods for LANs. This array of choices is good for the consumer and end-user but poses a difficult decision-making process for the network manager. . one has the whole spectrum of scalability from microcomputer PCs to supercomputers and some parallel processing. Case 14. Even supercomputers may not be adequate. The computer room was much quieter with Tricord servers running NetWare and various other PCs running e-mail and the communications programs. gateways and routers. The servers now had greater storage capacity and the system ran on a token ring topology reflecting the IBM heritage and the utility industry that EEI (Edison Electrical Institute) represented.5 gigabytes of disk storage running VM. multiple operating systems and many programming languages. What is discussed is middleware which is the software between the network and the application. And for software. Along with processors for telecommunications. we need hardware network connectivity hardware like bridges. This software with its APIs are especially necessary for distributed computing where there are a diverse set of platforms. It had a negative salvage value and thousands of dollars had to be paid to haul the equipment away. It cost $250 000 per year and had reached its cost-effective life. With the teleprocessing of large volumes of real-time data and multimedia traffic like video-on-demand. What had not been discussed thus far was software. (Norman and Tharnisch. together with middleware software and its many types. integrated or otherwise OO (Object-Oriented Systems) programming languages available to the application programmer in a distributed environment is summarized in Table 14. we now have APIs that are universally available to all applications for both client server and peer-to-peer interactions. We may also see ‘programming currently available hardware in the form of a distributed-memory multiple instructions stream multiple data stream computer . . 1993: p. Parallel processing and the emergent option in parallelism neural computing may become desirable. a multicomputer’. there is the choice of integrated packages with comms software. Some software like operating systems software and applications software and software needed at the connectivity devices are not discussed and considered beyond the scope of this book. there are topologies. MANs and WANS. but we also need protocols and a network architecture. there are choices in the US besides the OSI model and its protocols which has many supporters in Europe.1: Replacement of mainframes at EEI On 2 March 1995 the IBM mainframe 4381 was replaced along with its 7.

1: Top world telecommunications equipment manufacturers in 1994 Vendor ranking Alcatel AT&T Motorola Siemens Ericson NEC Northern Telecom Fujitsu Nokla Bosch Revenues (US $ billions) 15.) and Btrieve (Btrieve Technologies Inc. 4. Middleware’s next step: enterprise-wide applications. (1995). Norman. King. (1992). Java tools get real. 17(16). R. 77 83. F. 41 45.87 4. avoid duplicate mailings.41 12. M. R. 23(7). Data Communications. 181 214. (1992). (1992). Linthicum. InfoWorld.78 9. Serving up Apps.A. PC Magazine.65 9.Telecommunications and networks running on VM and rewrote them in the programming languages Magic (from Magic Software Enterprise Ltd. 12(1). ACM Computing Surveys. PC Magazine. 65 71. The new ‘LANbased applications allowed EEI to improve service to its members. p. 89 99. Source: p. 55 58.68 3. Personnel training. and Max. R. P.R. (1995). Bibliography Bernstein.08 8. Middleware demystified. 21(9). 86 88. 69 76. LAN 121(2).L. Data Communications. Messaging middleware: the next generation. Y. DeBoever. 14.V. Supplement 14. 23(1). Middleware: a model for distributed systems services. Making the most of middleware. Do we still need comm software? PC Magazine. Wadsworth. Models of machines and computation for mapping in multicomputers. Zahedi. Intelligent Information Systems for Business: Expert Systems with Neural Networks. P. L. Oct. 25(3). 205 236. 24(12). 14(18). and increase the quality of its databases’. 263 302.92 3. 201 204. F. (1995). 39(2). Datamation. 64.23 Source: International Herald Tribune. D. and Thanisch. Schreiber.28 13.) on the LAN. Middleware.J Jr. Dragen. 1995. 14(5). (1996). S.94 14.G. (1995). (1933). B.R. Rao. (1997). Database. M. and Peterson. Artificial neural networks: an emerging new technique. 21(3).S. (1994). Derfler. 160 . 157 164. L. Yoon. (1997). (1993). Communications of the ACM. Dolgicer. Data Communications. April 17.S. Rose. 1995. Data Communications.

There may well be a seachange. There are questions that arise such as: Where is the information highway headed? Will it be a ‘free’ way or a toll highway? Will it use the analogue telephone or the digital system or both? Will the use of the infrastructure be controlled and regulated by the government? What will be the role of the government in relation to the 161 . The impact will be profound but unlike the building of the interstate highway. It requires an investment that no one person or firm can afford but it is built and supported by the government and its services are offered to its users free or for a price. the information highway may not be so well planned or orderly. you could stop for a rest and a picnic on rest areas located strategically on most freeways. you would get off the freeway on one of the many exits where hotels and motels and restaurants have arisen to tempt the weary traveller. George H. Health-care. This freeway system has created new businesses and eliminated others. water or sewage.15 NATIONAL INFORMATION INFRASTRUCTURE One of the greatest contributions technological development can make is in systematically. advances in technology require an infrastructure that is very expensive but which has the potential of having a profound impact on society. It has encouraged travelling for vacations and business and in many ways has changed the life and landscape of the country. On this information highway you can cruise for business. Andy Reinhardt. The infrastructure highway enabled one to travel from most points to other points in the US without stopping for a single traffic light or pay a toll. intentionally building a national infrastructure for information exchange. it has made towns out of ghost towns along old highways and created new towns around the new freeways. every few decades. entertainment (including films and movies) and knowledge. Heilmeister Many of the technologies and players needed to conduct the information infrastructure are already in place. 1994 Introduction Every country and even many cities have an infrastructure. carefully. And if you wanted to rest for the night. adventure. You may no longer need to go to library buildings but can download books and articles that you want. life will be different with an information infrastructure. that for information. In short. the autobahn in Germany and the motorway in the UK. And now comes the possibility of another infrastructure. In fact. pleasure. It may not create (or destroy) towns and cities but they will change if people take to teleworking and reduce the need for high rise buildings and parking lots. One such infrastructure was the building of the interstate freeway system in the US under the Highway Act of 1956. electricity. which may have an equal if not greater impact on our lives. And some three decades later. education and shopping could all be electronic and a very different experience. Every so often. be it for roads. It may well be like the free-for-all competitive struggle of construction of railroads in the 19th century with highly competitive companies (and international alliances) striving to get advantages and be the best and first to succeed. the effects of the interstate highway are still to be seen. It was 41 000 miles long and cost over $50 billion by the time it was completed in the 1970s. If you got tired on the way.

Instead he inherited the phone that was in the apartment but also inherited a phone number 162 . We will then examine the advantages. Its current fibre backbone already serves intercity and intracity communications for professional and business applications. GSM (Global Systems for Mobile Communications) . TERA (Trans-European Trunked Radio) incorporates the best in trunk technology for standards for voice and data communication in optimized packet data services and for private mobile radio use. we take an overview of what is happening in constructing NIIs around the world. It is being said that what Japan will have in 2015 is a very large and beautiful lake but no fish to inhabit it nor any boats or recreation facilities to take advantage of its capacity. It started as a computerized telephone book in Paris. But Japan does not have a large cable market. VSATS (Very Small Aperture Terminals) is concerned with devices that appear on roof tops and behind garage doors have been agreed upon and now in the testing stage. Thus Minitel now has the same access to homes as did the telephone.1 NII related projects in Europe DECT (Digital European Cordless Communication) are standards being developed for the cordless telephone. IMPACT (Information Market Policy Action) is an effort to establish services in the areas of interactive multimedia and geographic information. RACE (Research and Development in Advanced Communication in Europe) is focused on integrated broadband communications and image/data communications.? What applications will be transported? Will there be universal access or universal service or both? Will privacy be secured and property rights protected? What will the system cost? Who will pay for it? When and how will it be implemented? In this chapter we will attempt to answer the above questions. The phone system was archaic. France then took the bold step to leap-frog existing technology and go for the next generation. One is Japan where NTT. In France. We will rely on the opinions of experts and chief executives in the field of telecommunications. It calculated (not allowing for inflation) that the cost of a terminal would be less than a Parisian telephone book and so gave every one a terminal with the ability to access the telephone directory. Now Minitel offers other services including video text and is an excellent base for an NII. Japan’s largest common carrier.8 Ghz band. A committee is now studying devices for satellite news-gathering systems. Singapore has declared the national objective of being a wired island. that belonged to the occupant some twenty years ago. Neighbouring countries like Hong Kong and Singapore are upgrading their telecommunications capabilities. GEN (Global European Network) is expected to be absorbed in METRAN (Managed European Transmission Network) which supports transmission across Europe at rates up to 155 Mbps. But first. there is Bete (Broadband Exchange over Trans-Eurpoean Links) with research facilities connected in France and Switzerland. plans a $410 Action Next Generation Communications infrastructure with fibre backbone and ISDN by 2015.Telecommunications and networks businesses and industries involved? Who will be the major users of the infrastructure. but there are many countries that are laying a foundation for NIIs. We will start by providing a scenario of what an NII (National Information Infrastructure) may look like. . We then look in greater detail at plans in the US. nor does it have the software or applications for the consumer and home market. 1994). networking and computing (Pelton. Germany has installed fibre in about 80 large cities and this is the basis of VBN (Vernittelandes Table 15. limitaions and obstacles for its implementation. a complete communication system for teleconferencing. OPN (Open Network Provision) is intended to make liberalization possible by eliminating the technical obstacles in the way of the deadline for the privatization of telecoms in the year 2000. There is also Brehan. A third generation European standard planned for 2002. Twelve years and about 6000 pages later. ESPIRIT (European Strategic Planning for Research in Information Technology) is concerned among other things with compression techniques for interactive media. NIIs around the world We do not yet have an NII anywhere in the world. The best known project is Minitel. Your author recalls wanting a phone connection in Paris and was asked to wait for three years. the standard will be DCS 1800 for 1. . cordless switchboards and office networks. video transmission. LAN interconnection and circuit emulation.

And so was born the ARPANET which was the test bed for many of the topologies and switching methods that we use today. community centres. Alongside network products were being developed. text. access for the poor. And then there are projects in the US itself between US companies and by US companies themselves. For the remaining part of this chapter we will look at plans for an NII in the US. its limitations and the obstacles in its path.e. at least not when it comes to implementation of the infrastructure.National information infrastructure Breitbandnetz) with broadband connections for video-conferencing via satellite with connections to international video-conferencing networks. owned by 12 electric companies that piggybacks the fibre optic network on the power grid. These are listed in Table 15. Not by design but by accident: an unintended consequence of the cold war. large and very effective like those implemented by DEC. It may have to settle for universal access (i. a successor to the ARPANET. minorities and the economically deprived) to the NII. Who are the players in this business of the information highway infrastructure? One may be tempted to say ‘the government’ since the government is often responsible for infrastructures. Even so. BERKOM (Berlin Kommunikation) is a pilot project that has applications for business as well as for publishing and medicine. It must bring creativity to the problems of supporting universal service. there are many projects with participation from Europe and the US including PEAN (Pan American ATM Network) with AT&T as the American participating with 18 other countries in a high speed network for data. Some of the private networks are international. 1994: 9 208). With the cold war between the US and the USSR heating up. But not so in the case of the information highway in the US. Also. switches. even in the US. We also saw a number of networks being implemented. which included hubs. One nation-wide project is Energis. businesses and homes in something like the scenario depicted in Figure 15. The Internet in the mid-1990s became the de facto national and even an international network. The government responded in the mid-1990s when it announced goals for connecting all schools. This liberalization (like that in the UK) freed the creative juices of the industry. the government wanted the system to be safe from criminals and abuse and it wanted the right to tap the system for what it may consider security reasons. LAN connectivity products and networking software (Data Communications. video and image transmission. examine its need. medical facilities. allowing it to go into computing and at the same time allowing computer companies to go into telecommunications. some public and some private but all from private enterprise. In addition to participation with countries in Europe. voice. routers. We shall discuss the Internet in Chapter 20 and the global network in the next chapter. the Court Judgement in 1982 which broke up the monopolistic telecommunications industry in the US. In the UK.1. the American Department of Defense wanted a communication system that would not be paralysed in the event of unfriendly action and at the same time a network systems that will enable its researchers in academia and research institutions to be able to communicate between themselves. Does universal services and universal access mean that the poor and the disadvantaged had 163 NII in the US The US has been on the path of an NII for the last thirty years or so. But perhaps the biggest milestone in the path towards an NII is . BT is also replacing all its copper wire with high capacity fibre. industry would rather have the government participate actively because the government has the power of regulating the industry. There are other test beds including one by BT in Ipswich. And then there were developments in architecture and protocols giving us SNA and TCP/IP. with cable companies making half their money from telephony. as in the case of road highways. Then there were networks built by individuals in the computing profession (with many contributions like CERN from Europe) that gave us the Internet. disadvantaged.1. one will find a very special case when compared to many PT&T countries in Europe: it has a liberalized telecommunications industry since 1981 allowing the telephone and TV on the same network and making fibre optics profitable for private companies in the UK. There are many projects with joint participation of many countries in Europe. libraries. Texas Instruments and IBM. Governments must choose points of intervention carefully. There was also another condition: the access to computing had to be ‘fair and equitable’ and essential services were to be available universally.

entertainment.4 Gbps channels using packet transfer mode. access to an encyclopedia or video-on-demand. publishing. game producers and software houses. which could be very expensive. The US in the early 1990s had a president and vice-president who did know the difference between a potato chip and a computer chip. including those of advertising. This would require a vast library available electronically at all times. And there was support in the legislature too. will the government do that or must the industry bear that burden? This is not explicit but there is support for the NII at high levels. to be subsidized? And if so. Or would consumers be happy with a limited set of films say the top ten ranked at the time? This would be relatively easy to provide. The war is ON. Watching the rise of NIIs are industries that stand to gain from an infrastructure. SMART VALLEY To promote data superhighway through private and public partnership. Even the multibillion dollar software companies are interested in developing software that will give access to the NII and the global network with a click of the mouse and. XIWT To work out the kinks in bringing gigabit technology to the business desktop and to the home. Each has billions invested in their technology and are pushing it hard for national adoption. and increasing trade. And who is involved in industry? There are of course the hardware manufacturers that gain in producing the many computers. MAGIC Employs SONNET links and ATM to create a gigabit WAN. Some of the pilot studies testing these preferences are listed in Table 15. 27). All industries look to an NII for increasing 164 . But the biggest players are the carriers of the gigabits of information that will roar down the information highway. banking. So the vibrations are good but they are in a cloud of uncertainty that makes the industry cautious and nervous. test beds are being used to determine consumer preferences. with a few more clicks.Telecommunications and networks Medical facilities Community Centres Schools NII Libraries Homes Offices Businesses Services Government agencies Figure 15.1 A scenario for an NII productivity. They are the carriers of telephone. NECTAR Uses ATM to link gigabit LANs and WANs to one another and to supercomputers. Since the stakes and risks are high.2 Pilot studies in the US FEDERALLY FUNDED PROJECTS AURORA Has 2. the leader of the House of Representatives in a TV address to the nation had a strand of fibre optics in one hand and compared it to the copper bulky cable that now connects much of the capital and called for modernization of telecommunications. 1994: p. interfaces and interconnecting products. microprocessors. In the early 1990s. frame relay and ATM. The software industry in the US should not be discounted for it accounts for twice what is spent on hardware. for each film could be run at ten minute Table 15. BLANCA Test bed links for FDDI LAN with SONET-based ATM switches. improving competitiveness. They each even had an e-mail address. An example of the type of question being tested is the consumer demand for any film at any time. VISTANET Concentrates on medical imaging with gigabit networks. FDDI. which by one estimate will be around $700 800 billion (Pelton. creating new products and services. COMMERCIALLY FUNDED COLLABORATORY ON INFORMATION INFRASTRUCTURE Attempts to find solutions to practical problems of end-user interface and network navigation NIXT Prototyping data highway concepts using the Internet.2. TV and cable. CASA Uses fibre optic lines at 633 Mbps for remote supercomputer use.

National information infrastructure intervals and all ten films shown on ten different channels, so that there would never be more than ten minutes to wait. And then, of course, a large number of permutations are possible. Selecting the right option in the correct mix is a serious problem, but a managerial one not a technological one. This is not to say that there are no serious technological problems. There are many and the most basic is the selection between telephone, cable and TV as the medium for information transfer. The telephone has the advantage of being the oldest with an entr´e into many million homes. e But the telephone has a heavy bag of disadvantages. One is that there is a heavy investment in the analogue equipment when the world is going digital. Also telephone systems are based on copper wire, a $60 billion investment, when fibre has a much greater capacity (for the same weight and volume) and is easier to maintain. Copper cables can be converted to fibre but at a high conversion cost. TV has many advantages, one being its entry into most homes in the country by being the most popular home appliance for a very long time. And it promises the choice of 500 channels. But TV has a serious problem for telecommunications: the image on the screen. The quality of the image is not so good on the screen for data and text as it is for pictures and it is text that will occupy a large part of the traffic in the future especially with the recent exponential growth of e-mail. Cable has a much greater capacity than the telephone, but it is mostly a one way street, what is scheduled (being determined by management of the cable company) with little if any interaction by the consumer. You cannot interact by asking questions about a product and placing an order. Cable firms are in general financially weaker and are heavily indebted. But they have the great advantage that with the addition of a box they can upgrade their system to pump torrents of digital data in addition to its images. Video-on-demand (one in every five people in the US rent a film video once a month) and interactive video games are also possible but all these enhancements will cost more, around $250 350 per home. Neither cable nor TV has the nifty switching facilities or the communication skills and experience of the technical personnel of the telephone companies. It would seem that cable and TV have many characteristics that are complementary to the telephone and that they should cooperate and produce a better joint product and better services. Instead it seems likely in the highly competitive world of the US that these industries will compete in each other’s markets. This may well expand the market but this may still be somewhat of a zerosum game where one’s gain is the other’s loss. The winner in the final analysis may not win on technological grounds but on how they finance expensive projects and how they package their technology with a content that the end-user wants to consume. The technology that the end-user may well want is at least one channel that is ubiquitous, multimedia, interactive and end-user friendly. The experience for the user should not be difficult or frustrating; in fact, it should be an adventure and even some fun. But what mix of technology and content will the consumer prefer? ‘What is truly impossible to foretell is how much ordinary people will pay for the new offerings . . . But at present, no firm quite knows what people are willing to pay for, nor how to deliver it to them at an attractive price.’ (Economist, Feb. 25, 1995, p. 63).). A comparison of telephone, cable and TV for important technological characteristics is shown in Table 15.3. These carriers are testing their packages of technology and content (like news, weather maps, shopping, etc.) in different test beds across the nation. The decision on which package to offer will not be made mostly on technological grounds but on consumer response. If you give the consumer a choice, you will get a response in the market-place. Testing this response are numerous pilot studies, a selection of them listed in Table 15.4. Some of these projects have since been abandoned and others may be on the way to being abandoned. So there is a high risk because the

Table 15.3 Comparison of telephone, cable and TV Cable Affordability Availability Bandwidth Acting Capacity Content Ease of use Security Reliability Popularity Interactivity Fairly good Good Good OK Very high Questionable Very good Not so good Not so good Not much Not possible Telephone Good Very good Very poor Excellent Not high Depends on user Excellent Poor High High High TV Fairly good Excellent Good OK High Regulated somewhat Very good Not so good High High High


Telecommunications and networks
Table 15.4 Sample trials for interactive services on TV Year 1994 1994 1994 1994 1994 1995 1995 Location Florida Omaha New Jersey New York New Jersey California Washington Company Time Warner US West Bell Atlantic Nyex Bell Atlantic Viacom TCI Targeted end-users in 1995 4,000 40,000 7,000 800 7,000 Not available 2,000

funding involved for each player is not just in the billions but tens of billions of dollars. And so the carriers are looking for partners in this adventure of high risk and high profits. Also, management is concerned about control so that they may protect their existence and survival. One example is the publishing industry. They recognize that the telecommunications technology can soon transfer a 20 volume encyclopedia in a matter of seconds. So why would anyone own a set of encyclopedias or even buy books or go to the marbled library building when they can scan any book, decide if it is worth reading, and then download it from a national digitized library, all the while sitting comfortably at home? The same logic applies to films. Why go to a cinema when one can see a film-on-demand? And so the moguls of the film industry in Hollywood are concerned. They have the content but not always the control over the distribution of the content of the highway. Thus there has arisen a set of incredibly complex potential alliances of computer manufacturers, software houses, carriers, publishers and Hollywood studios. Often one firm is with another in one alliance and competing against the same firm in another

alliance. What is at stake is the confluence of the telecommunications industry, the computer industry, the publishing industry and the film industry. Each player has to decide that mix of content and distribution that they want to control, all this within a cloud of uncertainty of governmental intervention and consumer response. One cannot push the government lest they make too many regulations, but one can test the consumer response. The main options are compared in Table 15.5. They are being tested in different pilot studies, some of which are listed in Table 15.2. In these projects, what is being tested is not just technology and content but also the managerial skills and viability of alliances with a mix of partners. There have been failures of alliances, which may partly be due to the stock prices and foreign exchange rates dropping, but may also be due to the difficulties of managing complexity of international multi-industrial alliances. This may have been the reason for the loss of $3 billion by Sony in its purchase of the Hollywood giant Columbia Pictures, and the $7 billion sale of the film company MCA by the Japanese computer giant Matsuhita. At the time of the purchase in 1990, Matsuhita called the MCA purchase a pillar in the multimedia strategy for providing entertainment software that could use electronic hardware. Significant mergers and alliances are not always with content as witnessed by the BT (British Telecommunications) which is making strategic acquisitions and alliances in the telecommunications and computer industry not just in Europe but in the US. The search for the right alliance and the proper mix of services offered with adequate control was on and most likely will go on right through the 1990s.

Table 15.5 Alternative strategies and their characteristics for NII Cable Topology: Protocols: Backbone: Orientation: Key Users: Relationship: Interactivity: Unswitched Proprietary analogue Analogue, fibre, satellite Entertainment Homes One-to-many None Telephone Star or circuit switched. Good switching capability ATM, ISDN Fibre optic Communication (verbal) Everybody One-to-one Some Internet Packet switched, router TCP/IP NSFnet Communication (written) Business, government and individuals Many-to-many Some


National information infrastructure

Issues for NII
There are many issue involved, both technological and organizational. One technological issue concerns the choice of a network architecture and set of protocols. In the US there is strong competition between SNA and TCP/IP. The former has the clout of its designer IBM, but the latter has the support of the operating system UNIX and the reality is that it is adopted by the ubiquitous Internet. In Europe they do not have such a difficult choice. Being the home of international organizations for standards it has the OSI by ISO. Europe also has an adoption of the B-ISDN. There is also the problem of bandwidth management and determining what bandwidth will be required and where. Who will want interactive computing and who will not? Who will use information for consumption and who will want it for communication? Or both, and then what would be the mix? Another technological decision relates to having an ATM that can allocate bandwidth on demand, handle interactive multimedia in large quantities at high speeds, and assign priorities to data even at the cell level and all without any loss of content. As for organizational issues, there is the problem of determining who controls the content of what is transmitted. Surely those who own the films will want its control but how about the control of crime in films? How about pornography? And how about the content of what an individual user can send and receive on the NII? Can someone put a computer program on the net that he or she has written even if designed to break a government code? Each of these issues is contentious and is being contested in the courts. Many of these issues are addressed in the US Telecommunications Act of 1996 along with the provision of a V chip to provide parents with control on what their children can and cannot see. But these control issues of the 1996 Act are being hotly contested in the courts. Industry wants government to be interventionist but only in selected areas like the hooking up of medical specialists with rural clinics; the connection of schools in rural areas to educational institutions; and the connection of poor communities to community centres and national libraries. Surely the government has some role to play but a balance has to be found between control of information and entertainment and its freedom of expression. How much protection should the

couch potato be given if he (or she) wished to see mud wrestling all day long? Or gamble half the day away? Whatever the decisions, whether they be technological, organizational, or political, they should result in the NII being open, easy to use, affordable even for the poor, multipurpose, seamless in accessibility and protective of privacy, as well as having a rich and balanced information content. Governments and industries should move towards common standards that are not too rigid, so that technology can grow and innovate without one segment getting an advantage over another. We conclude this chapter with a positive note from a CEO in the communications industry: ‘We are poised on the verge of an information infrastructure that I believe will serve us well in the next century. Its development is inevitable.’ (Heilmier, 1993: p. 34).

Summary and conclusions
We started this chapter by comparing the interstate highway in the US with the information highway infrastructure. It may be no coincidence that the information infrastructure is called a highway because there is a desire for it to be as successful as the interstate highway was and still is. There are of course commonalties, but there are also differences. The interstate highway was ordained and paid for by the government. It had origins and destinations in a static system that is bounded by geography. It had a grand design. In contrast, the information highway is not ordained by law or the government but may well evolve through private enterprise and populous support. The infrastructure for the information highway has origins but no destination and no assurance of where it is going or what traffic mix it will carry. It has no specific design except that there must be a seamless dynamic web of networks that will be equipped with intelligent ATMs that will direct (switch) the multimedia traffic roaring down at gigabits per second to where it needs to go without any loss of content integrity, privacy or security. For local connectivity we may have local telecommunication networks in addition to some private networks, LANs, as well as wire-less and wired telephones. The NII that will emerge may well not be a monolithic system but many highways all interconnected seamlessly together, but with gigabits per second speeds and even tera bits per second 167

Telecommunications and networks capacity. Also, the system will most likely be open with universal or near universal protocols and access. Ideally, we may even hope for the integration of 6GCS (sixth-generation computer system) with communication carriers, suppliers, as well as with the electronics industry to offer content that is rich and varied. An NII may lead the country to greater productivity and even greater competitiveness in business and industry. It may create more jobs and may improve our health services, enhance our knowledge, and give us greater access to entertainment. But if this mix of services is used mostly for playing games or entertaining the couch potato, then the information infrastructure would have failed. It would also fail if it did not meet the real-world needs of its end-users and citizenry, or if the system was too expensive or too complex to use. The infrastructure would also be a failure if it was inequitable and left out those who could not access the infrastructure, be they children, business people, civil servants or blue collar workers. It is here that government can play a part and subsidize for services that give access to schools, community centres, and hospitals that would otherwise be inaccessible. The government may also subsidize pilot demonstration projects that are too risky for private enterprise, thereby providing important feedback on end-user needs and acceptance thresholds. This would help raise the levels of interoperability and end-user friendliness of interfaces, and prepare the public for the changes in technology that may well affect the way they work and play. Furthermore, the government should not withdraw its support for free enterprise in computing or support monopolies or the separation of the telecommunications and computer industries as happened in earlier days in the US. Standards will be crucial, not just to an NII nationally, but to it internationally, enabling it to be connected globally. This is the subject of our next chapter. Concert which is a result of a joint venture with MFS Communications in the US. VIAG Intercom will use its 4000 kilometres of fibre optics owned by a regional power company that VIAG owns. MFS Communications in the US has won a limited licence to operate a fibre network in Paris, the carrier will free users from having to purchase leased lines from France Telecom. Telecom Finland Ltd has signed an international agreement with MFS to operate an ATM backbone extending to various European cities including St Petersburg, Russia. MFS is operating a similar network (as with Telecom Finland) in Frankfurt, Germany, to link with an international backbone to networks in London, Stockholm and the US. MFS has won a international carrier licence to own an end-to-end subsea fibre between the US and Sweden. Source: Data Communications, Feb. 1995, p. 18, and July 1995, p. 48.

Case 15.2: Share of the European VAN market
VAN, Value-Added-Network, includes e-mail, EDI, reservation transactions, videotex, card authorization, and enhanced fax. In Europe, these VAN services are dominated by PT&Ts as shown in their share of the world market: France Telecom 20% DBP Telecom 13% Telefonica 10% Reuters 7% BT 4% IBM 4% Swift 3% Unisource 3% Stet 3% Geis 3% Others 31% Source: Ovum Ltd (London), printed in Data Communications, Aug. 1993, p. 18.

Case 15.1: Alliances and mergers between carriers
British Telecommunications PLC plans an alliance with VIAG Intercom AG in Germany aiming to compete with Deutsch Telecom. The joint venture will offer domestic and international private virtual networks for voice and data traffic with international connections to be handled by 168

Case 15.3: Telecommunications law in the US
In 1995, the telecommunications industry in the US had revenues of over $700 billion,

National information infrastructure around 6% of the entire US economy. The growth came despite it being burdened under the Federal Communication Act of 1934. At that time there were only 3 TV-cable companies and one telephone monopoly. In the 1980s, there was some deregulation with the dismantling of the one monopolistic AT&T into seven large local telephone companies. Then came 400 longdistance telephone companies (dropping rates by 65%); numerous TV and cable operators; 30 communications utilities; the increase in our daily lives of computing, information processing and satellite services; a proliferation of PCs; use of digitized voice-mail and the merging of data, text and voice transmissions; and the popularity of e-mail and the Internet. Such changes in environment required a change of the 1934 Act, which resulted in the Telecommunications Act signed in February 1996. The 1996 Act is a comprehensive rewrite of regulations regarding the communications and computer industries, including the deregulation of long-distance and local telephone companies, cable operators and computer manufacturers, allowing each to compete on each other’s otherwise protected turf. The 1996 Act allows for cable rates to be gradually deregulated by 1999 (allowing the high rates in the short run to provide capital for integration and expansion), and also extends the limits on the number of TV stations allowed in a single company from 25 to 35. Between 1934 and 1995, the FCC (Federal Communications Commission) supervised the monopolistic carriers. With the 1996 Act, the FCC will encourage creativity and a free market economy for communications and computer industries whilst insisting that all the competitors will be interconnected and will share certain resources and facilities. One problem is that long-distance companies do not have the wires into homes and must rely on local providers. This reliance represents 45% of total long-distance revenues. Economies can be achieved by integrating long-distance and shortdistance transmission resulting in efficiencies and cost reductions that could (theoretically) be passed on to the consumer. The contrasting philosophical approach of the 1996 Act compared to the 1934 Act is that in 1934 the government tried to predict all the possible future problems and formulate regulations for each of them; the 1996 Act recognizes that the telecommunications and related industries are too volatile and shifting to enable prediction of the future and so the industries are allowed to develop and innovate unfettered with detailed regulations to be legislated as and when the need may arise. The Act has already resulted in a flurry of acquisitions, joint ventures, alliances, partnerships and mergers in attempts to ‘bundle’ services of the residential phone with long-distance calling, TV, cable, fax, video, wireless and cellular communications, paging, on-line information services, Internet services, entertainment like video-on-demand, and interactive services. Conceivably, most if not all services will be available with ‘one-stop-shopping’ through one carrier or utility in one bundled package and charged for in one monthly bill.1 The 1996 Act will not formally institute an NII or even a formal national infrastructure in the US. However, one stated objectives of public policy is to have computer access to all homes, libraries and health-care facilities, especially in the poor and rural areas, by the end of this century. The FCC has been empowered to encourage computer access to all schools, libraries and health-care facilities, especially those in rural and isolated areas, through inducements and incentives to the appropriate industries. It is expected that the 1996 Act will encourage investors, reduce the inhibitions of entrepreneurs threatened by regulations, offer open markets resulting in reduced rates, and spur innovation that will bring new products and services to the market-place, thus giving consumers greater communication choices and rate options. The Act will trigger an explosion in learning by people of all ages, at all income levels, and in all areas (urban and rural). It is also expected that soon medical doctors and other professionals will be able to offer multimedia state-of-theart services and consulting advice to all persons who need it irrespective of where they are. For the telecommunications industry, the Act allows integration, mergers and expansion. For the consumer, they will have integrated and end-user friendly services at a lower price (it is predicted that phone prices will drop by 20% in three years). There are two additional features of the 1996 Act that were extensions to the 1934 Act. These relate to the V-chip and the display of indecent material on the Internet. The V-chip provision requires that all TV equipment manufacturers install a microcircuit (in all its equipment with 13 inches (or more) of 169

B2. This will not rule out challenges in the courts. p. 2. 27 32. Overview of enterprise network developments.H. sex. R. (1994). Lippis. IEEE Communications Magazine. Telecommunications in Europe: creating new links. abortion. E. 30 (1). broadcast programme Rating Boards are expected to assign a ‘grade’ classifying each of their programmes based on violence and ‘indecent’ content. Telecommunications. Reinhardt. D. The term ‘indecent’ is open to interpretation and it is feared that it could well include anything on.. the first network architecture. Time. (1995). 170 Supplement 15. Planned date for a nation-wide fibre infrastructure in the US. X. Feb. SNA. 2. 19 (3). 30 37. Jackson. and even crime. 3. is concerned that the 1996 Act may not ‘prevent domination’ by a few large corporations over ‘what is rapidly becoming the central factor of American life’. Internet. McCarroo. 2015 2015 Bibliography Benhamou. International Herald Tribune. 23 24. 27 32. Taylor. Modified final judgement and the break-up and deregulation of AT&T. Kay. which allows freedom of expression. Feb. (1994). To help parents make the blocking decisions. the dawn of networking for the masses. The indecent provision makes it criminal to ‘knowingly’ publish material that is ‘indecent for minors’ on public networks like the Internet. Sources: 1.S. A15. (1993) Strategic technology for the next ten years. the first LAN standard. Deregulation of the telecommunications industry in the UK.R. 30 37 Pelton. Telecommunications. Data Communications. IEEE Communications. 28 (12). Battle for remote control. Creating a European information infrastructure. The Coalition argues that the law will chill free speech such that public discussion would be diluted to the level of that which is acceptable only to children. R. 23 (17) 60 64. A. (1994). Telecommunications. and Woodbury. British Telecom buys 20% of MCI and the rush to merge and acquire telecommunication assets starts globally.N. p. the first public networking service.Telecommunications and networks their diagonal screen) costing about $1 each in order to enable parents to ‘block’ material that they consider unsuitable for children. Byte. J. CIO survey on the national information infrastructure. 1996.3 While the resolution of the legal battle over the 1996 Act is still uncertain. Gopher at://gopher. including the 1984 decree to break-up AT&T. (1996). M. violence and terrorism. G. The NII: more than just a data superhighway. The new public network. Judge Greene. Wall Street Journal. 13 24. 47 48. birth of networks. 28 (1). (1996). (1994) NII development: where do we go from here.A. N. 69 72. obscenity.25. the American Civil Liberties Union and 19 other organizations filed a suit challenging the law as being unconstitutional by violating the First Amendment. Violators of this provision face a fine of $25 000 for individuals and $500 000 for companies in addition to a possible jail sentence of up to two years. nudity. Also. there is less uncertainty about the economic consequences: the restructuring of the communications industry and the likely dominance of the US in the global telecommunications /faq. Building the data highway. including some 150 international Web sites by turning their pages black in mourning for 48 hours. Telecommunications. 1996. 49 national organizations (including one in New Zealand) organized the ‘Coalition to Stop Net Censorship’. 14 October. K. say. Heilmier. The term ‘indecent’ is to be defined by the FCC in the ‘public interest’. Mercer. 1993. The day the bill was signed into law as the 1996 Act. 34 (1).1: Milestones towards the development of an NII 1966 1972 1976 1978 1980 1982 1985 1993 ARPANET. 31 (2). 12. 46 74.2 Other organizations also protested. Spring. Ressner. T.panix. known as the telecommunications Czar in the US for having made many important rulings concerning communications. ETHERNET. . rape. (1994). 28 (1). Planned completion of ‘Next Generation Communications’ infrastructure in Japan. The Washington Post. J.

and the brain drain resulting from global outsourcing (contracting of computing services) abroad. We will make no value judgements. In this chapter we will be concerned only with applications that have an important global dimension. that of global outsourcing. We could get sick in one country and have lab results sent to another country for an international expert’s opinion. as it is not yet safe or secure for monetary transactions. Wayne Masden Introduction In earlier discussions. as well as privacy and security considerations. We could order a rare book. All these problems are magnified in global networks. We will examine the Internet in great detail in another chapter and we will see that it was not the consequence of any grandiose careful design or any design at all. however. namely transborder flow of data. And yet. . In addition. transborder flow of information. It certainly was not expected to carry the traffic for businesses. at least a de facto global network: the Internet. and since we do not have any NIIs it would be logical to conclude that we do not have a global network. there are problems unique to globalization of telecommunications like international standards. We have e-mail. First. the protection of intellectual property and the viability of a secure means of making transactions on the global network. we could cruise the souks of Istanbul and the bazaars of Bombay. it has no formal organization for its maintenance and control. We could dropin the Victoria and Albert Museum in London or the Louvre in Paris for an interactive look while still sitting in our chairs in Australia or New Zealand. There are also problems of global telecommunications relating to developing countries. However. control of ‘content’. If we had. though you can do some business on the Internet and can order flowers in one country to be delivered in another. An NII is a subset of a global network. Also. 171 . a best seller or a professional book available anywhere in the world and read it in the comfort of our homes. we will take an overview look at global networks. . For example. we do. you would be wise not to make any large financial transaction on the Internet. What would that do to global levels of education and development? What would that do to our paradigm of higher education? How will telecommunications affect the development of developing countries? Will it improve communications and bring the world together or will there be a gap between the information rich and the information poor? How will that affect our life-style and standard of living? We shall examine these implications and the applications of telecommunications in four chapters later in this book. customs duty or immigration officers. We examine one important consequence of global networks. the solution of the international protection of intellectual property rights is a prerequisite to having free access to books in a digital library.16 GLOBAL NETWORKS The world has become a market-place for information. and then discuss some outstanding issues in global networks. international protection of intellectual property. Computerized data recognizes no border check-points. we mentioned or implied the many problems like bandwidth capacity limitations. It is these subjects that we will address in this chapter. we are more interested in the technological feasibility and prerequisites of these applications. but we do not have universally accepted e-money (electronic money). These are some of the topics that we shall examine in this chapter.

and about 150 with at least e-mail services through IP or via more limited forms of connectivity. In Thailand. a trained cadre of technical personnel. along with the rise of multimedia traffic across their national borders. More specifically. The deregulation of AT&T (in the US) in the 1980s saw the rise of MCI and Sprint and the plummeting of rates and increase in services in the US. and experience in computing. that have a liberalized telecommunications industry. In Hungary. poverty and debt payments. There were in mid-1990s. Many of these countries in the 1990s are seeing the convergence of the industries of telecommunications. Some countries think that free distribution of information can be more detrimental than Technological infrastructure What is needed for a telecommunications infrastructure is a reliable technological infrastructure that includes a uninterrupted power supply. control of content. The technological infrastructure includes equipment for telecommunications. the prerequisites include: technological infrastructure. A physical telecommunications systems is necessary so that it can be coupled with at least grassroots connectivity like the Fidonet. For some countries.Telecommunications and networks Global networks A global network connects PCs. where timely data is crucial. Not all countries recognize that privatization is important or even relevant for them. This list implies that there are prerequisites for a successful global network. This happened with the telephone industry in France in the 1970s and the 1980s. and because there is no global predictable legal and regulatory framework for fair competition and incentives for investment in telecommunications and networking. the declared desire is to go digital. They may have to build and maintain a national backbone and help provide gateways for other nations. there are only two. the UK and the US. clean water and shelter. satellite and microwave links. but the ‘killer’ applications are e-mail and messaging. Their existence varies greatly in different countries with varying priorities. because there is little well trained personnel (administrative and end-users) to absorb the technology. the drive is towards an advanced cellular digital communication system. which has simple protocol for storing and forwarding messages and a strong error checking mechanism for ensuring correct transmissions. Many countries attach low priority to privatization and claim that attending to the more basic needs like food and hunger. government support. as with the Adameus reservation centre in Germany. over 70 countries with full TCP/IP Internet connectivity. information and entertainment. because of the absence of consistent end-user interfaces. When a country goes from little or no infrastructure to the latest technology. are more important that the distribution of data and information. Poor countries with high unemployment assign a very low priority to labour saving devices and computing and telecommunications fall into that category. It must also observe working hours that allow for the different time zones and be familiar with the different languages used. Government support The success factor for any information systems application is the support of top corporate management. Though there are some 40 countries that have some privatization and liberalization. backwardness may be an asset. the emphasis is on cellular phones. solution of international issues. And in China. For global systems it is necessary to have the support of the government and its PT&Ts where much of the power resides. because of the narrow spectrum of services offered globally. International issues An organization with a global network must understand and follow the many different regulatory standards concerning transmission of data across national borders. In Vietnam 172 . These will now be discussed below. The attitude of the governments relating to open systems and privatization of the telecommunications industry is important. and international standards. it is estimated that they need over $50 billion including the costs for an increase in its telephones by some 80 million. workstations and data centres around the world through fibre optic. In Russia. Global systems for information systems processing is not yet viable partly because of the lack of standards in important areas. there are $40 billion in projects to lay fibre optics. Some applications are mission critical. it will leap-frog entire stages of development.

a process that is slow and frustrating at best. can come only after international standards are agreed to and this is where the bottleneck often appears. Other countries acknowledge the need for free information flow for their industrialization and development and may even help them to leap-frog stages of development. Consequences of global networks The satisfaction of prerequisites for global networks can result in many applications. To them. data pertaining to economic potential and structure is a national resource and that their governments should be able to exercise control over its collection. some applications cannot become common until some outstanding issues concerning global networks are resolved. 173 Control of content Many countries question the value of the content of much that may roar down the global networks. These are very enduser friendly. These are discussed in four chapters in the next part of this book. Likewise we cannot see businesses using global networks very much unless data and funds can be transferred securely and safely across national borders.Global networks beneficial to their environment. and so prices cited by information providers may vary up to tenfold across countries. We start our discussion with the consequences of telecommunications on developing countries. we cannot have the advantages of digitized library (despite what this could mean to education and development) without resolving the problem of protection of intellectual property. They do not necessarily want to be influenced by the life-styles and cultural values of the sending country. many countries have found a balance in favour of at least privatization and universal access despite its PT&Ts having the monopoly of 100% of its voice service market. the advantage of 2000 Newsgroups or 500 channels is irrelevant when they have a very high signal-to-noise ratio anyway and bandwidth is very expensive. ‘Many countries believe that . They argue that telecommunications are important to high tech personnel but of marginal value to countries with low tech people where ‘small is beautiful’. Important consequences of global networks are: the impact of telecommunications on developing countries. Other examples in Asia are Indonesia and Malaysia. use and distribution. so that it can be read by anyone on the Internet irrespective of their hardware and OS configuration. 1991: p. Getting different countries to agree to international standards is often not a technological but a political issue which requires the harmonization of conflicting national policies and local regulations. with Australia and New Zealand coming close to being in the zone of ‘no-regulation’. Standards Acceptance of international standards is important for interoperability of telecommunication devices and hence interconnectivity. global outsourcing. especially the free movement of pornographic or violent material. A good example of this is Singapore that has made many moves towards making the Internet more accessible to its citizens despite its strong concerns about the importation of pornography. transborder flow and the protection of intellectual property. The monopoly of telecom companies often give them power over pricing. In Pakistan and the Philippines prices for telecommunications are raised in order to attract foreign exchange. However. The most common application globally is e-mail and file transfer. This is an old argument that has been around the UNESCO for years and runs into the philosophical argument of censorship and freedom of speech and expression and the right of a government to determine what is best for its population. The acceptance. Europe is not too far behind despite the slowness in the adoption of the European Commissions proposals to pull down barriers to privatization and liberalization of their telecoms. Despite the contradictory advantages and disadvantages of regulation and privatization.’ (Woody et al. however. For example. They also argue that telecommunications will produce advertisements for products of developed countries which not all countries can afford or need.. Putting a ‘home-page’ on the Internet requires the use of a simple language called HTML which defines the look of the page. These issues will be examined later in this chapter. 36).

(Sadowsy. electronic fund transfer. Telecommunications also facilitates all business dealings and their conclusion through EFT. However. One example is financial services. The process of entering the global community is as much one of entering a new culture as it is one of learning about and exploiting the new set of technologies . 46). 1993: p. future generations of information services and products that live on the network. This is possible by software that can prevent material with obvious screen names and a history of sending certain types of material. Also. in 1995. The government was caught by surprise and may well prevent that from happening again. even if this means access to an information-rich environment. To achieve electronic communication. are not likely to find prenet analogues as easily. As with security. Traditionally. the contact was through mail. These basic needs have a higher priority than the desire for interactive communications. This is partly between administrators and managers in developing countries who once studied in developed countries and can still benefit greatly by maintaining communications with their contacts. the developing country must have a basic telecommunications infrastructure. To understand the electronic network culture and be most effective in exploiting it. Some countries may also want to restrict information that they consider harmful to their society or may object to material that they consider subversive. Controlling content is also a problem in developed countries where parents want to control the material that their children see on the Internet. The need for a telecommunications structure cannot always be justified on the platform of communications. Not all developing countries want unfettered communications on the Internet for various reasons. This infrastructure is capital intensive and may not be affordable by countries that have to consider the needs of foreign exchange for more mundane but essential imports for daily living like food and medical supplies. Now communication with experts. Another very important use of telecommunications is for faster and more reliable communications through e-mail and file transfer using global networks like the Internet. for some developing countries this may result in a negative balance of payments because the imports may tend to be larger than the exports resulting from telecommunications. connectivity between critical network links. . This infrastructure is partly technology which will include availability of adequate and reliable bandwidth. teachers and professional mentors in developed countries. telephone and even diplomatic pouch. if the imports (of raw materials and capital goods) were judiciously selected then they could improve productivity and gross national product in the long run. consultants and professional mentors can be fast and even interactive or at least on-line on the personal and professional levels. Developing countries could also benefit from businesses and industries now possible because of telecommunications. one must live in and explore it. A 1996 Telecommunications Law in the US requires 174 . there are many consultants in developed countries who worked in developing countries and have much knowledge and even dedication to these countries but all this is lost because to a lack of communication. interconnectivity and better navigational tools for the Internet. These contacts can refer the query to others when necessary and provide information of the latest state-of-the-art technology. Some do not want material that they consider pornographic and what developed countries may consider artistic. . But at best this is slow (even on the phone where getting an international connection is not fast or easy even after allowing for changes in time) and not always reliable. Now these consultants can keep in touch with their contacts in developing countries by giving advice and even downloading materials and computer programs on the Internet as and when needed.Telecommunications and networks Telecommunications and developing countries Telecommunications can help developing countries (and developed countries) through increased trade including that with multinational companies. supply of spare parts and access to maintenance services. But there may be other surprises in store. you can build a ‘firewall’ but there are many creative people around that want to break that wall for purposes that may be ideological or even just the fun of breaking the rules. a fairly uninterruptible power supply. reliable telecommunication media. However. For example. Chinese dissidents in the US sent a sea of anti-governmental material on the Internet on the occasion of the anniversary of the Tiananmen Square incident. including intelligent user agents and knowbots.

disadvantages and limitations involved. These include having access to a pool of professionals in developing countries without having to pay generous fringe benefits or having to contend with the unions of developed countries. changing national laws regarding foreign investments and national political ideologies that are sometimes hostile to collaborations with foreigners. Good programmers and analysts are not the preserve of developed countries. However. The MIPS (millions of instructions per second) cost to labour cost ratio in a developing country is almost 10 50 times that of a developed countries (Lu and Farrell. One American hiring programmers in Austria and Eastern Europe in the 1970s was asked why he was looking outside America when Americans had the best departments of computer science and were the leading manufacturers of computer equipment. Moreover. However. these costs may drop over time partly because of the large increasing pool of computer specialists. locate and retrieve information over the world-wide global networks. Countering the advantages of global outsourcing are the many risks. The other reason for the high MIPS cost ratio in developing countries is that the cost of labour (denominator of ratio) is low in developing countries even for skilled computer specialists. process and time of completion. 293). many of the professionals that are involved in outsourcing of computer services are family businesses with little or no experience of large complex information systems development. are technically trained that speak English and are ‘abundantly available at very low salaries’ (Apte. Also. Discriminating information coming from MIT (Massachusetts Institute of Technology) as to whether it is technological or ideological would be difficult. But controlling content world-wide even with international agreements may be very difficult. The most obvious advantage of global outsourcing is the cost advantage. There is also the problem of finding appropriately trained information systems personnel with an adequate infrastructure of telecommunications and other required computing resources. There are also additional costs of telecommunications. In the early days of outsourcing. 290). time zones. Also. But data collection has since been largely automated. It is the extra additional advantages and limitations that is the subject of rest of this section. travel and the costs of training vendor personnel to the organizational and national culture of the host country. Changing from the typewriter to the PC for daily communications requires a sea change in attitudes and life-style. This is partly because the cost of computing equipment required (numerator of ratio) is high in developing countries and because of the high cost of transportation. 175 Global outsourcing Outsourcing is the subcontracting of services. you need the equipment to send and receive electronic information. The answer was that many programmers outside America have fewer resources of computer memory and computer run time and so they have to work harder and write more efficient programs. and the negotiation of legal contracts for the outsourcing. Yet all the advantages. there are over a quarter of a million new graduates per year that . There are other advantages of global outsourcing of computer development projects by developed countries. The electronic infrastructure also includes the knowledge and ability to navigate. In India alone. The risks include political and social instability of the vendor country. This statement may well be true of many a developing country in the 1990s. The disadvantages of global outsourcing are that in developing countries it requires a much greater degree of monitoring (and control) of quality. And when they do. import tariffs and the many middle men involved. the need for outsourcing has not diminished since it has now shifted to programming. they do not even have typewriters on every desk. However. In many developing countries. There could be potential problems with foreign exchange. it is very much part of their way of communicating. global outsourcing offers knowledge and access to entry into the market of the vendor country. In some developed countries there is a PC on most office desks and homes and the typewriter is no longer in local production. risks and limitations of outsourcing within national borders applies to outsourcing abroad plus much more. In such services the developing countries have a comparative advantage. 1990: p. it was mainly in the preparation of data. 1990: p.Global networks a V-chip that allows control over content. there may be resistance among systems professional in the developed countries.

Large masses of people in counties like the former USSR. One important obstacle in global outsourcing to developing countries is the lack of experience in many types of information systems development projects in developing countries. which only increases the brain-drain problem. Given all the drawbacks and limitations of global outsourcing. Akinlade of Nigeria.1 Types of outsourcing 176 . large. The developed country does not always discourage this brain drain because in many cases they pay for the education through scholarships and jobs. Some developed countries do not help the problem by liberal immigration policies. by working on real-life. . it is still attractive to many developed countries to have their software constructed abroad. One can expect global outsourcing to grow vigorously in the late 1990s and early 21st century. the theory that he has learned . his working environment does not motivate him to play an active part in the country’s evolution of this very exciting discipline.1. the software engineer in a developing country may never be able to put into practice.Telecommunications and networks especially in times of recession when outsourcing personnel may become potential competitors. argue the developing countries. including networking software and even all network services. What greatly inhibits greater global outsourcing are the constraints in transborder flow and the lack of protection of intellectual property. This. . They go to the developed countries to meet their clients in outsourcing and are then hired by their clients. A skilled programmer or analyst can get an immigration visa to the US. complex systems. There is already a brain drain resulting from students of developing countries that do not return from developed countries where they go for their advanced education. The professionals involved contract Raw data F I R M are mostly educated in the developing countries and trained on local problems. 71 2). To add to this. has the following observation: In a typical developing country. there is the brain drain resulting from global outsourcing. There is one important problem arising partly from global outsourcing: it causes a brain drain from the vendor country to the developed country. These topics we will now discuss. . but for the developing countries they are crucial for their economic development. and there are firms in the US that aggressively import computer programmers and analysts and facilitate their immigration. several application generators are available so that developing non-trivial software systems from scratch is not only unnecessary but is also uncommon . (Akinlade 1990: pp. . Transborder flow Changes in the world political scene offer opportunities for trade and communications across national borders. The basic players and content of global outsourcing are summarized in Figure 16. For the developed countries these resources are relatively marginal. is unfair and even an act of stealing by the developing countries. Eastern Europe and China have joined market Machine readable data Program specification OUTSOURCING Working programs VENDOR Specification for services Services delivered Figure 16.

Every year billions of dollars’ worth of software is pirated and development managers must continually try to secure their systems against such piracy. computer and communications services affecting transborder data flows. (Gassman. Since transborder flow is the concern of multiple countries. promote access to data and information and related services. Information systems will share data across the many barriers of borders and geography whilst overcoming the risks of foreign exchange and national political stability. seek transparency in regulations and policies relating to information. 4. If these rights are respected by the important trading countries. security safeguards. still problems posed by cultural differences between managers. 3. Firms must also be concerned with their proprietary software despite the GATT agreement of 1993 94 that protects copyrights of software. The threat of espionage from the former Soviet countries may have decreased but international sabotage is still a threat. 4. develop common approaches to dealing with issues related to transborder data flows. There is a story that the software bought by Iraq for the control of its air defence system contained a computer routine that was activated remotely and destroyed the Iraqi control and command system just before the Gulf War. And so the need for security is a very important concern of all managers of development. 204): 1. data quality. are now more open and less afraid of foreign domination. 2. that were market-oriented economies. 2. 8. Sixteen out of the 24 members of the OECD countries have passed guidelines in 1988 which state eight principles (Gassman. not just the concern for international violations of security but even violations of a firm’s database. developing countries may export (by design or accident) computer viruses across national frontiers. however. in which they agreed to: 1. openness. and avoid creation of unjustified barriers to the international exchange of data and information. use limitation. Telecommunications and networking could bring instant international communications across national borders though there may still be restrictions on transferred flow of data and information. 3. Also. consider possible implications for other countries when dealing with issues related to transborder flows. the 24 countries of the OECD (developed countries) adopted a Declaration on Transborder Data Flows. This is not only of concern to multinational firms but all firms that have secret and confidential data and are vulnerable to unauthorized access. and. But it does raise grave possibilities such as the ‘bugging’ of a country’s software system by its enemies or even terrorists. to develop harmonized solution.Global networks economies. accountability. 204) Part of the problem of seamless transborder data flow is the concern for the security and privacy of data that flows across national borders. If this story is true it may never be confirmed. 1992: p. but unfriendly to Western companies in computing. International telecommunications has helped global outsourcing to the advantage of developed countries for it has made developed countries more accessible. Such a virus made in Pakistan devastated billions of bytes of data all over the US. Furthermore. In 1985. 5. when appropriate. end-users and developers and the wage differential between different countries. Other countries. There are. it has been the subject of discussions in international bodies. 1992: p. purpose specification. because there is some uncertainty about the observance and enforcement of the protection of intellectual rights. collection limitation. telecommunications across the oceans has encouraged unauthorized access to computer systems such as the espionage for the former Soviet KGB by hackers in West Germany. The GATT agreement of 1994 has reduced trade barriers and acknowledged intellectual rights of the computing industry. 6. This growth will come partly from multinational companies that will prosper and the international firm will be the norm rather than the exception. 7. There is also the danger that espionage will shift from the political to the economic stage. But there is a downside to it: developed counties are attracting computer personnel from the developing countries and so there is a brain-drain in developing countries. then this contributes towards the opening of new markets of information-based services and will lead to globalization of IT. like India. 177 . individual participation.

Outside the government. estimates that 90% of all software in use in Asian countries is pirated (Weisband and Goodman. The BSA (Business Software Alliance). Software Publishers Association in the US. For some. the US has agreements with many countries. FAST. the temptation to pirate software is great. CAAST. If the effort at the international and bilateral levels fails. It is estimated that in 1993 the US software industry lost $2. The other problem is that software piracy is often not considered wrong or ethical. the agreement with South Korea took four years before the government brought their first copyright action. This action against Cyprus is intended to be a strong signal that the US will no longer allow intellectual piracy which has existed with impunity for many years. Software piracy is the unauthorized copying of computer programs for the purpose of selling the illegal copies or for unauthorized commercial or even private use. there are associations for software protection like the BSA. What makes software piracy different from other thefts is that its occurrence is not detectable or obvious as with say the theft of jewels. INdian FAST in India.1 billion in Japan and about $10 12 billion world-wide in developing countries because of software piracy (Gwynne. the US demanded the strengthening of the protection of intellectual property lacking under the multilateral Berne Convention. 87). and this was an important objective of the Uruguay round of GATT in 1993. is somewhat inadequate or even in doubt. A copyright under the Copyright Act of 1976 (in the US) classifies computer programs and data as non-dramatic literature works deserving copyright protection. trade associations like the International Intellectual Property Alliance and the BSA based in Washington have their own crusade against software piracy.Telecommunications and networks Protection of intellectual property Transfer of technology can be inhibited by the lack of protection for intellectual property which includes patents. there was no mechanism to enforce the copyright protection law agreed to in June 1992. but not all of these agreements have been enforced. copyrights and trade secrets. In our world of ever-changing computer technology which emerges and becomes . operating in the US. The developed countries. in 1993. It is proceeding at all levels: international. 1992: p. 101). it is like driving a car at 60 or even 80 in a 55 m.p. For example. But there are countries outside the GATT agreement and the adequate enforcement of the agreement. It was achieved and over a hundred countries have signed the agreement. 178 At the international level. Section 301 of the US Trade Act was amended to toughen penalties for countries that do not protect American intellectual property rights.h. zone. Then. From the point of view of the management of development of information systems. Thus far the attempt at incarnation is confined to individuals. In the US. Whatever the motivation or justification. 1990: p. software piracy is cheap. a source of great friction between the two countries between 1994 and 1996. they have identified Ong Scow Peng based in Singapore with illegal programs worth several million dollars. easy and efficient. the US government can act unilaterally.73 billion in 1989 to around $340 billion by 1996 (Schware. even by a software industry that acts independently of its country. however. or offer pirated software free as an incentive to buy hardware. Canadian Alliance Against Software Theft in Canada. For example. the US government suspended the duty-free status for Cyprus on the grounds that Cyprus had imported almost four million blank cassettes. There are ways to protect unauthorized copying but there are also ways that the pirate can get around it by opening up the programs and rewriting them. can be quite harmful to developing countries. Programs that would sell for $600 are being sold by software pirates for as little as $15 in shopping malls in Hong Kong and Seoul. and INFAST. copyrights are more important. Thus there is little incentive for customers to buy programs legally constructed by software houses in developed countries. The US is the source of much new and innovative software and is very concerned. At the bilateral level. In China. 15). Federation Against Software Theft in Germany. 1992: p. bilateral and unilateral. from the point of view of developed countries. possibly to copy and then export both software and other intellectual property like films and music. but a national incarnation. Some think that it is a transfer of technology to developing countries that is long overdue. object to the ‘free ride’ especially when the software production world-wide is expected to grow from $36. The stated goal of BSA for Mr Ong is ‘incarnation’. Also. Computer dealers download unauthorized copies of software to the hard disks of computers that they sell.

We already transfer funds by EFT (electronic fund transfer) but that is between banks. Another approach is ‘money tokens’ which are prepaid smart cards. so why not make money while we can. The greatest interest to business lies in the confluence of PCs. But paraphrasing John Maynard Keynes. For example. safely. But. there has to be a way to transfer funds globally. Businesses face computer fraud and customers are afraid of giving their credit card numbers on the Internet. there is a consortium of large electronic firms called CommerceNet which has $12 million in state and federal funds that is working towards making the on-line business easier and safer to use. the customer gets confirmation messages via the Internet and receives the goods within two days if in the US. there are problems with shopping in the electronic shopping mall. Various approaches to e-money are being pursued by firms like DigiCash in Holland and CyberCash. there is the netiquette that disciplines those that take advantage of the system. Smart cards with stronger safeguards against misuse are being developed by Mondex. an American firm working out of Austria. a consortium of British banks. And there are other ways to make money: doing business on the Internet. spamming (flooding the . E-money can be called digital money because the code for the money and the parties involved are often digitized. commerce and money. system with the same posting) is looked down upon as is the regular repetition of the same blatant advertisement. But in the long run. But this may take time before the system is thoroughly debugged and certified as easy and safe to use The customer can access the system. that is to be a store of 179 Global network and business The combination of processing text and images on the Internet makes it a good target for use by business. The electronic payment system may make it easier for money launderers and tax dodgers. Meanwhile. Software houses and credit companies are now working on ways to encode credit card numbers on-line without any danger of misuse.internet. If an order is placed.Global networks obsolescent very rapidly. It will also increase outsourcing by developed countries for they will be less afraid of their software and data/knowledge being pirated. is changing all the time and any details stated here need to be updated to be valid anymore. as well as the protocols on the Internet. The concept behind e-money is to provide an electronic signature (like a watermark on some bank notes) with the recipient checking with the issuer. Thus. The customer can then access the ISN server and browse through the ISN catalogue which has up to 20 000 computer hardware and software products. by entering a command as simple as http://shop. in the long run. the issues raised are not just for banks. electronic money. A business can pay a fee of around a thousand dollars a year and have a ‘home-page’ of advertising material that is theoretically accessed by around 25 million people connected by over 20 000 computer networks run by universities. Credit cards are such a vehicle but transferring funds by credit card globally on the Internet is asking for trouble. control of software piracy is good news for developing countries. So. It will increase investment and licensing by developed countries and will strengthen the computer industry (especially the software industry) of developing countries. Each issuance is identified uniquely as is a bank note by a serial number. not much harm can be done for the copied signature is only good for one transaction and large transactions will be verified before being accepted. developing countries will benefit. that can be transferred electronically. the software pirates may say: in the long run we are all dead and gone anyway. If copied. But what if you wish to spend more than what you prepaid for? What is needed is e-money. businesses and governments. netiquette. for business to benefit globally from this confluence and global telecommunications. for example the ISN (Internet Shopping Network). However. The actions by the US (governmental and private) is very bad news for developing countries for they may now have to pay for the software (and copied hardware) that they use. However. quickly and cheaply. What is needed is a means of transferring money by individuals to any other party. it is important for developing countries to have good cooperative relationships with the leaders of computer technology in developed countries. A secure transfer system must be able to confirm tens and hundreds of transactions a day. Digital money does provide a medium of exchange but it does not solve the second role of money. but for national agencies trying to prevent such practices especially from going across their borders. To solve some of these problems.

retailers. Finally. and where power politics complicate negotiations. You will use . Actually this is what CyberCash proposes to do. 1994: p. fast food restaurants. transport commodities. the number of applications is expanding. At each stage of the trade cycle. for what rate? Can digital money be used for collateral for a loan? Can it be exchanged for foreign currency and. design and manufacture products. take loans. In addition. checking the cards on the spot to confirm that the merchant took only the amount that you had planned to spend. Guarding paper money is expensive. Should the equivalent of real money earn interest and if so. You’ll download money from the safety of your electronic cottage. there is also the possibility of the electronic wallet or electronic purse. The sum will be automatically debited from your account into the merchant’s. (Levy. at what rates? Answers to these questions now rest with national central banks and national monetary policies. against it. For this we need transborder flow of information. if so. many transactions today are conducted with digital money if you consider all the EFT transfers which are digitized in crypotographically sealed streams of bits of data moving between government clearing houses. Digital money can be refused as legal tender. data are recorded. The memories of George Orwell’s Big Brother looking over your shoulder immediately comes to mind. Privacy is under threat. issues of global telecommunications are resolved in bilateral and international forums where world views and national interests continually shift.Telecommunications and networks value. Credit cards are also popular but not secure against fraud or violations of privacy. school cafeterias and places where cash is now essential. helping to define markets. Summary and conclusions Information and telecommunications technologies play an important role in world trade. the understandable desire for security against fraud. In our traditional monetary system. This approach will protect some (especially the money launderer even across the borders) but will conflict with the government’s desire to control such illegal transfers. 180 What some people think we need is digital money. There is also a conflict between the needs of financial institutions like banks and individuals. convenience stores. To be competitive in world markets business people in all countries need access to information technologies. What is needed is not just a ‘digital signature’ which requires establishing authenticity that can be traced back for undesired purposes. IT managers must remain alert . people deposit money in a bank and. The problem is one of balancing the legitimate needs of privacy protection against violations of privacy and potential surveillance. ATM (automatic teller machines) are popular but not always safe from muggers. with output delivered to managers. which is a palm-held calculator-sized reader of smart cards that can be used at petrol stations. What is not digitized satisfactorily yet is the last mile in the money transactions. that of payments in cash. Digital money has a demand because traditional money as bills can be copied by high quality copiers. National governments may well hesitate to abdicate these decisions to cyberspace or cyberpunks. toll booths. Individuals are concerned with privacy and security being compromised. sell goods. The problem with credit cards is that they make it possible to construct a digital profile of the customer and trace a person’s card transactions with ‘decidedly discomforting intimacy’. and safeguarding the national monetary system against fraud and misuse as in money laundering. A definitive statement of the issues of global telecommunications may never fully be achieved because information and telecommunications technologies are continually changing through research and development. . disgorging them whenever you spend money. Actually. They will hold real money in escrow for its equivalent in digital money. For this to happen with digital money. and exchange payments. 176). a secure means of monetary transactions and protection of intellectual property. workers and clients largely by telecommunications. . as well as electronic wallet. but a ‘blind signature’ that is easy to use and has the elegance of anonymity of paying cash and yet retaining the store of value. draw credit or even demand cash. while banks are concerned with a system that is cheap and fast to use for transactions across borders. But this raises other questions. stored and processed by IT. cards in telephones (including those in the home). In addition to e-money and digital money. we will need to have legal money in the real economy deposited for each unit of digital money.

And the PTT alliances now being struck throughout Europe may make it even tougher for providers of managed network services to offer international outsourcing.Global networks to these interests and power shifts and adapt to them. It serves more that 104 000 PCs through its two large centres in England and the US and through 10 outsourced access networks at BT/Concert (UK). 75 80. Iceland Air Case 16. 80). Amadeus owns and manages most of the backbone equipment at the two centres in Atlanta (US) and London (UK). There will be more confrontations like the threat by the US to impose a stiff tariff on selected goods worth over a $1 billion on China if China continued to refuse to satisfactorily protect American intellectual property rights including those of computer programs. 1994: p.25 networks in Belgium and the UK. They are configured with uninterruptible power supply and routine monitoring of traffic coordinated with the telecommunications and processing centre in Germany which is equipped with a cluster of IBM mainframe computers. Finland.’ Telstra in 181 . (Goodman et al. Case 16.1: Global outsourcing at Amadeus Amadeus Global Travel Distribution SA (Madrid) is a computerized reservation system with its central processing done at Helder. including international data/voice network solutions. in Reykjavik. There is a high correlation between poor telecommunications and many developing countries. car rentals. 1994. Air France in Valbounne. The consolidation will include the assertion of their international rights. France. Data Communications. French Travel industry. Spain. In addition it serves airlines of SAS in Scandinavia. Germany. Germany. There was a last minute compromise and there will be more. The information infrastructure in developing countries is often inadequate for a robust and sustained support of networking activities. and Finn Air in Helsinki. hotels and other travel related items for agencies that own PCs. Global Outsourcing: What Works. Thai airlines in Bangkok. The developed countries will want to concentrate on integration and consolidation. Lufthansa airlines in Frankfurt.2: Telstra in Australia Telstra is the largest integrated telecommunications carrier in the Asia-Pacific region and ‘provides a comprehensive range of telecommunications solutions. Airinter in Paris. 1993: p. Both sides seem to be preparing for test of strength to see who blinks first. the Spanish travel industry. 43). X. Iceland. measuring availability from endto-end including access lines. global telecommunications will not suffer from such blinking competition and will continue to grow and prosper in the future. Source: Peter Heywod. These centres as well as the other networks offer services not just to the travel operators for airline tickets but also for reservations for theatres. Amadeus collects a small fee for each booking which in 1994 was around 135 million. not including the SLA (Service Level Agreement) in the contract so that no adaptation to changing conditions can be made without a battery of lawyers going into play.’ (Heywood. What Doesn’t.. network management and service performance. Hopefully. Evaluating the experience of running such a large outsourcing processing network the lessons to be learned were less technological and more of an organizational nature. Thailand. The 1990s will be a period of flux for international issues of IT. SITA access network. Nov. The developing countries will want to leap-frog the growth curves without giving up any of their existing privileges. Developing countries that have a weak capital structure also have weak planning for investment in infrastructure and a poor payoff for telecommunications. Electronic network connections have a potentially higher payoff for developing countries because of the lack of reliable alternative delivery mechanisms in the developing world. They are connected by separate 384 kbps transatlantic lines to Erding in Germany. pp. Iberia in Madrid. 21. Scandinavian travel industry. and the German travel industry. and recognizing that the ‘PTT infrastructure monopolies make outsourcing mission critical international backbones too risky at present. All countries may agree on the need for global telecommunications. France. but how they go about it will differ. especially as regards the protection of intellectual property including compute software. Such lessons were dividing the outsourcing between at least two operators so that the competition will drive prices down.

universal access to networks. 6.8 million and 8. Mexico. Telstra’s MobileNet digital network is one of the largest in the world in terms of geographic coverage. protection of intellectual property rights. Some of the reasons for the inability of developing countries to achieve telecommunications parity with developed countries are: ž old infrastructures. Columbia. 1995. the European Association of Manufacturers on Business Machines and Information Technology has a list of six principles for a GII (Global Information Infrastructure): 1. p. Oct. 16. The high income countries which account for 15% of the global population have 71% of the world’s telephone lines. access to research and development. low quality services and very little interoperability between systems. 182 In 1995.9 million main lines. 16. The analogue coverage is planned by the government to be phased out by the year 2000 and taken over by its digital system. 3. The GSM. the 24 countries of the OECD (Organization of Economic Cooperation and Development) generated 85% of the world’s telecommunications service revenues. Oct.4: Slouching towards a global network Guenther Moeller. a two-way paging and voice mail computer messaging service. They are already being installed in Brazil. ž know-how deficit. interoperable systems and applications. ‘Looking globally. It is expected that the majority of the 800 million subscribers in 2000 will have such a service. Without the infrastructure they cannot integrate into the global market. high tariffs and restrictive governmental policies. 19. Vietnam and Zambia.3: Telecom leap-frogging in developing countries ITU estimates that 4 billion out of the 5. Sri Lanka. new applications. Russia. Case 16. 5. Developing countries have a great fear that they will never catch up with developed countries. Finland. Global System for Mobile digital cellular system. The consortium is attempting to develop technology that will enable information systems that make up a network to collaborate more effectively with each other by integrating databases and providing services. Source: International Herald Tribune. is being installed in Eastern European countries and is planned for extension into 100 countries. Telstra is also part of the TINA consortium formed in 1993 with about 40 of the world’s major network operators as well as switching and computer manufacturers. Spain. ž ž ž red tape. 2. privacy and security. we see a patchwork of incompatible communications networks marked with high costs. including international roaming in 150 cities in 27 countries. Consequently they cannot afford the telecommunications infrastructure that they so desperately need. And without integration they cannot pay for the infrastructure a catch-22 situation. 5. a 64 kbps connection could cost around US $8000 with equipment costing around US $30 000. 4. 1995. and a fax and data service. the World Bank has estimated. Chile. What we . will require $55 billion (about 10% of the world’s annual spending on telecommunications) every year over a six-year period to finance the necessary development in developing countries that include the former Eastern Europe bloc. Ghana.Telecommunications and networks 1994 was the 16th largest telecommunications company in the world with a telecom revenue of US$ 9. Director General of EUROBIT. To overcome this disparity. p. and May 17. high costs: for example. 4. Malawi. lack of funds: not all developing countries reinvest their profits into the network but spend it elsewhere. One approach to leap-frogging the telecommunications gap is to use fixed wireless that provide customers service from a radio station to antennas in homes and offices which perform like regular phones. 1994 p. Telstra also has an analogue coverage which is one of the world’s most extensive. Germany.7 billion people in 1994 did not have the basic telephone services. Case 16. Source: International Herald Tribune. reaching 89% of Australia’s population.

3 Supplement 16.8 Index 5.5 0.8 4.3 6. where 10 is the best score.3 0. a telecompetitiveness index was calculated using 43 specific .5 2.3 4.3 3. 1995.4 7.8 1.5 8.8 Contd.0 11.3 12.2 3.6 7.4 2.9 17. German and US companies In 1995.3: Telecommunications media for selected countries in 1994 Country Argentina Australia Brazil Canada China Czech Republic France Germany Greece India Indonesia Israel Hong Kong Japan Korea (South) Phone lines PCs Cable TV (units per 100 people) 4.2 5.0 1.31 1.1 13.7 5.’ Source: International Herald Tribune.4 57.6 5.8 8.2: Index of global competitiveness Using 1991 Performance data.7 N.8 18.1 1.9 54. and Deutsche Telecom claims to have the most advanced ISDN.9 0.0 4.3 47. 8. 183 Large gainers from international pirated software as a percentage of their total software used: China Russia Thailand India and Pakistan 8 95 2 87 Source: Fortune.8 3.0 Quality Canada France Germany Japan Singapore UK USA 3. 1995. factors measuring critical areas of telecommunications.8 13. the densest network and the most extensive cable network in Europe.0 39.5 5.2 4. p.0 2. p.1 R&D 6.1 49.4 2.5: Alliance between French.8 6.6 4. a consortium between French Telecom. The alliance of these three large companies is expected to produce a serious competitor in the global telecommunications market.0 14.7 7.1 7. Thailand.9 6. A few such categories measured and the overall index of telecompetitiveness (TC) appear below: Infrastructure Productivity Penetration Canada France Germany Japan Singapore UK USA 7. especially China.2 Case 16. July 10.6 8.7 21.5 7.3 4.3 9.0 48.5 5.9 6.4 54. Supplement 16. 11.7 48.1 9. French Telecom is the telecommunications arm of the French nationalized PT&T.3 5.05 0.Global networks need is a common worldwide infrastructure to communicate information at reduced cost. Sprint is the third largest American long distance carrier. 121.0 7. Supplement 16.24 5.9 4. India and Pakistan 1.6 14. 1995.1: World-wide software piracy in 1994 Total losses to international piracy US $8.0 Source: Stentor.7 5.1 0.2 0.7 0.6 1. US Sprint and Deutsche Telecom was announced. Performance in 10 categories were measured and converted on a scale of 1 10..2 5.4 6.9 4.48 0.9 2. p.9 4.2 3.7 1.7 8.08 billion that include (in billions of US dollars): Japan US France UK and Ireland Others.A 6.3 39.7 2. Dec.3 6.3 26.4 11.3 20. Russia. Oct.

3 2. 44 50. 21. Global outsourcing of information systems and processing services.2 28.7 20.M. A. S. Journal of Systems Management. Derollepot. U.. 2(12). (1995). L. Ruth. 25(11). Levy. 9(4). 287 303. 59 66. Communications of the ACM. (1994). E.L. M. International carriers team up for global reach.2 47. 1 7. and Ruthowski.3 14. Data Communications. 12 17.7 1. 42(6). Data Communications. (1994).America W. Computer. Vol. Journal of Information Systems Management. 26 34. J. and Fleck. p. Personal Computer World. L.2 8.3 17. Communications of the ACM. The global diffusion of the Internet: patterns and problems. 95(1). H. U. 213 215. Technology Review. 7(3). (1993).G.3 68. Gupta. G. Liebermann. P. Datamation.3 2.2 37.0 4.1 1. Jan.A. International software piracy.8 2. Software piracy. The impact of transborder flow regulation. 174 177. 27 31. Woody.V.0 Source: Extracted from IEEE Spectrum.3 59.P.6 23. 32 35.R. Weisban. 15 16.3 15. E-money: that’s what I want.Europe None Urban & Rural None Urban & Rural Urban Limited Urban None Urban & Rural None Urban & Rural Urban Urban Urban Urban & Rural 184 .0 16.0 1. Website traffic management: Coping with success. (1994).1 48. (1994).P.4: Telecommunications end-user service available in regions of the world Lines COUNTRY Africa Australia Eastern Europe N.2 1. The Information Science. The struggle for global networks.7 9.2 10. Goodman. International Journal of Computer Applications for National Policies.1 0.9 32. Stalking Asian pirates. (1990). Denmead. 1996.G.2 1. 38(1).6 5. S. Sadowsky.1 29. and Seymore.7 9. Network connectivity for developing countries. R. 7. International telecommunications: the current environment. Controlling copyright infringements of intellectual property: Part 2.2 50. 23(17). Gassman. 70 73.9 35. Gwynne. (1990). 40. Y. (1992). 5(1). Global networks. 45(7). S. 87 90. (1992). 404 413. 36(8). Malhotra. (1997). 85 88. Supplement 16. 37(8).5 40.0 15. Journal of Systems Managment.Telecommunications and networks Country Malaysia Mexico Netherlands Portugal Russia Singapore Sweden Switzerland South Africa Taiwan Thailand Turkey UK USA Venezuela Phone lines PCs Cable TV (units per 100 people) 14. (1992). (1991).1 15.9 66.I.E. Press. P. 39(17). Strauss.America ATM None None Frame relay None None ISDN None None Private Urban Urban & Rural Rural Urban & Rural Urban & Rural Rural Urban Limited Urban None Limited Urban Pacific Rim Limited Urban Rest of Asia None S. S. (1992) Information technology developments and implications for national policies. F. Internet.9 3.5 Bibliography Apte. Guynes. 2(1). Wired. (1993). C.4 1. 42 47.


This Page Intentionally Left Blank .

it was desirable for all messaging in an enterprise to be integrated. we now have video-conferencing and multimedia conferencing. the messages were structured and often special forms were required. especially in business. money had to be transferred. Gasparro. . as well as teleconferencing. there was cooperative processing where telecommunications and its networking was used by people working together and sharing contributions to the same data/knowledge-bases. with payments consisting of data transfers between one machine and another. MHS is still evolving but some directions are becoming clear. Such messages were sent electronically by EDI. And whether collaborative or not. which is referred to as ‘snail mail’ by e-mail enthusiasts. The telephone was mostly a one-to-one relationship and. Approaches to such integration is referred to as the MHS. the Message Handling System. This required another set of messages to be transported and is known as Electronic Fund Transfer (EFT). 1993 Money is really information. Nor do we discuss e-mail which is electronic. This is known as electronic mail. that by knowledge workers and teleworkers. Electronic Data Interchange. These approaches to messaging had some overlap but were not mutually exclusive and so they coexisted but needed to be coordinated. but to conclude transactions. On the contrary. or e-mail for short. But traditional mail is slow and with computers it became possible to send mail electronically. it is too important and a discussion of it will make this chapter unduly long. But for the transaction to be completed. agreements have to be formalized in writing and then sent by mail. With telecommunications and networks business. the downsized data center may return as the full sized mail center. We shall examine these approaches in this chapter along with some of the messaging approaches mentioned above.17 MESSAGING AND RELATED APPLICATIONS Within three years. Therefore we shall defer the discussion of e-mail to Chapter 19 and discuss it in the context of its greatest use. we had teleconferencing and. Daniel M. We also defer the discussion of video-conferencing to the next chapter where we discuss it in the context of distributed multimedia processing. This is not because email is unimportant. James Martin Introduction Early messages were conveyed by telephone or by mail. E-mail was unstructured and unformatted for personal and informal mail but. and as such can reside in computer storage. with more advanced technology. transactions were conducted with the necessary messages transported swiftly between buyers and sellers wherever they might be and wherever the price was right. We will not discuss all types of messaging especially the telephone and the post office because they are not electronic. for business transactions. Funds were then transferred electronically between computer systems and over distances using telecommunications. when it had to be a many-tomany relationship. Conferencing is useful in bringing people together on-line and in real-time consultation. As e-mail goes mission critical. as distinct from all other mail. . 187 . Meanwhile. The coordination could done by groups in cooperative or collaborative processing as in a client server environment as discussed earlier. e-mail will become the critical application at virtually every Fortune 500 company.

Telecommunications and networks Teleconferencing Teleconferencing is a method of electronic communication that permits interactive exchange between two or more persons through their clients or terminals.1 Traditional messaging vs. invoices. EDI 188 . Teleconferencing eliminates this time lag since participants are waiting to respond to messages that are sent to the screens of linked workstations/desktop PCs. a message is delivered instantaneously to an electronic mailbox. or between computers in conference rooms as far apart as one or more continents. The EDI thus differs from e-mail which is primarily text that may or may not be formatted and structured. prepares order Routes invoice to shipping dept. Teleconferencing has been used extensively in discussing product development and product production. the investment in teleconferencing is often a business choice of preference. thereby having greatly reduced the product development life cycle. With electronic mail. bills of lading and related business document necessary to perform specific transactions. many professionals like the travel and believe that interpersonal relationships are important when conducting business. but there may be a delay before the intended recipient collects it. With early computer processing. However. The cost of teleconference equipment is high. while e-mail often includes personal correspondence EDI is designed for business information exchange. another disadvantage. Often the terminals of conference participants serve as electronic flip charts enabling the participants to access specially prepared conference materials from a database. However. Traditionally. Teleconferencing saves travel time and travel costs. The transactions are recorded on specific standardized forms that allow for specifying content as well as for checking of content and certain errors.1. Electronic data interchange (EDI) EDI is a direct exchange between computer-tocomputer of separate organizations of standards business documents such as purchase orders. Process of EDI The process of EDI as compared to traditional processing is shown in Figure 17. when the executive’s time is too valuable to be spent in air travel and frequent business trips over great distances are necessary. EDI Link Processes order Seller Policies + Procedures Buyer's Computer System Seller's Computer system Figure 17. Also. so face-to-face communication is preferred over teleconferencing. A teleconference program handles the logistics of this communication and determines who has the ‘floor’ when several users want to ‘speak’ at the same time. Some systems include a gag feature to prevent a single person from monopolizing communication channels. inquiries were made by telephone and then confirmed by mail. the paper hard-copy agreements were converted into machine readable form for computer processing and then the transaction Telephone Traditional Processing Modes BUYER SELLER MAIL EDI link Buyer Policies + Procedures Calculates inventory level Computes when reorder is needed Calculates reorder quantity Issues purchase order Sends items ordered + bill Shipping dept.

Thus the process is greatly automated. lower costs and reliability. This discussion enables us to now specify the conditions necessary for a successful EDI: The two trading partners must have compatible computer systems (hardware. Transaction documents and product identification must be standardized between the two trading partners for format. or the money is transferred between the banks of the buyer and seller through EFT. The necessary quantity to be ordered is specified in a purchase order and sent to the seller of the inventory ordered. then a direct link is possible as shown in Figure 17. however. standardization of document format. This may not seem to be a problem at first sight because most businesses have computers. many computer systems are not compatible in hardware and its operating systems software. Unfortunately. There must be an EDI transmission link. an accounts receivable is issued and sent to the billing department which then collects the money. The buyer’s computer systems governed by policies and procedures of the buyer (as embodied in a computer program) determines when to buy. sorting and forwarding messages. The third party intermediary function can cause extra cost and time and so large businesses demand that their supplier have compatible computer systems and large suppliers wanting the business will comply. Mailbox facilities must be available for storing. If their systems are compatible. This determination is made by an inventory control computer program which determines the reorder point given expected demand and inventory-on-hand. the process is greatly automated eliminating the conversion from hardcopy. the standardization of EDI 189 . EDI between two trading partners does require that the trading partners have computers and a communication EDI link. Automation also reduces transcription and conversion errors that would otherwise occur during the manual conversion of data from hard-copy to machine readable form.2 Alternative EDI links was consumed. Standardization We have mentioned standardization on two occasions: one. They then go through a third party provider that serves as a go-between clearing-house converting the transaction from one system to being acceptable to the other systems. At the same time. The seller’s computer system acts on the order by rules embedded in its computer program. and two. sorted and sent when desired to the destination. Such a system can give the seller a comparative advantage over a seller that does not have EDI and could ‘lock’ the customer for convenience.Messaging and related applications EDI link THIRD PARTY EDI PROVIDER EDI link CUSTOMER Direct EDI Link VENDER CUSTOMER Figure 17.2. This third party can also serve as a store-and-forward point as does a post office. The speed of processing can reduce the inventory otherwise held by the buyer and reduce inventory holding costs. The order is processed and instructions sent to the shipping department which then ships the necessary goods to the buyer. may be by a cheque mailed to the seller. software and processing procedures). either direct or through a third-party provider. Under EDI. reducing costs of data entry (and correction of errors that do occur) and of processing as well as the time required for both data entry and processing. where messages are stored.

there are two competing standards. Giant computers in banks. This includes format for purchase order forms and packaging slips. European firms doing business with Americans must follow the X23 standard. Electronic transfer of funds In 1965. This is important as trade and commerce increase and becomes more global. Global standards other than for EDI are necessary not just for global outsourcing but also for transborder flow of data and the globalization of IT. In contrast. A summary of the above discussion on EDI appears in Table 17. and attempts to maximize the role of the private sector. either directly as with compatible systems or through a third-party EDI provider Both parties must have a ‘mail-box’ capability Both parties must follow standards for document format EDI transmission ADVANTAGES OF EDI (over traditional modes of telephone and mail) Fast Lower costs of data entry and processing Certain operations are automated Errors especially in data entry are reduced Reliable and well established system There are many spill-over services including those in outsourcing DISADVANTAGES AND LIMITATIONS Closed system Not universally used Real-time but can be ‘slow’ Cost for software installation computer output became the input for another computer system. thereby eliminating the costs of reentry of data which is an important component of the total costs of processing. it was found that over 25% costs of processing was for reentry and that over 70% of 190 Table 17. Initially. a tool of industrial policy. Whilst EDIFACT is attempting to make its standards international with or without American support. with massive memories.1 Summary on EDI PREREQUISITES OF EDI Both trading partners have computer systems The two computer systems must be linked. where EDI is at around one-fifth of the level of the US. made the following prediction:Ł In our lifetime we may see electronic transactions virtually eliminate the need for cash. which covers over 80 business or transaction sets that are mostly generic with some that are industry specific (Trauth and Thomas. In one study in the eastern states of the US. Thomas J. Despite the problems of standardization. But as multinational trade in IT increases between developed and developing countries. For the X23. . To draw down or add to his balance.1. With the increase in global outsourcing and multinational firms. it was the exchange within a firm. Commerce and Transport’). systematic. pragmatic. the need for compromise may move us towards a truly universal international standard. entrepreneurial and individualistic. anticipatory. Watson. there were around 5000 EDI sites in 1990. It is very different in approach and philosophy to the X23. For an international EDI. 1993). In contrast. reactive. EDI will be used increasingly to automate transactions between computers without human intervention and the output of one application for one trading partner becomes the input for another trading partner. the exchange concerns are global. the customer Ł Reprinted with permission from IBM. The other type of standardization. will contain individual customer accounts.Telecommunications and networks transmission. as well as being responsive to governmental direction and national policy. In Europe. The former is usually settled between firms and corporations that are trading partners. the X23 is distributed. EDIFACT is centralized. EDI is often associated with business transactions between buyer and seller. But there are many other transactions that are performed by EDI. EDI has great potential both nationally and globally. 1993). that for EDI transmission. is more difficult because it involves not just telecommunications vendors but also many countries when international commerce is involved (Gordon. then chairman of IBM. there is the more recent European standard. One is the X23 developed by ANSI (American National Standards Institute) in the US. international standards could be only a guide whilst the EDIFACT is a single European standard adopted by all European trading partners. EDIFACT (FACT is the acronym of ‘For Administration.

to seven-day period between the 191 . Billions of pounds sterling are moved daily from one set of accounts to another using computers and telecommunications without any currency exchange or paper to record and process transactions. the amount he punches out will move out of his account and enter another. one check.Messaging and related applications Point-of-sale terminal Point-of-sale terminal Authentication Bank database Bank1 computer ACH database Bank2 database Authentication Bank2 computer Bank1 Automatic clearing house Bank2 Transactions Service Transactions Service Customern Customern Figure 17. millions of times each day. paper-free banking is still far from being realized. amount. one piece of paper. However. depositor. Information on cheques (payee. now occurs. In the private sector. is not physically moved from one location to another. Instantaneously. (See Figure 17. such as the transfer of funds for government employees’ life insurance programs.) The US Treasury Department uses EFT for recurring payments. Billions ‘change hands without the use of one pen. punch out the transaction figures on the terminal’s keyboard. hundreds of thousands. account number. A network of terminals and memories extend across city and state lines for electronic funds transfer (EFT). although we have not altogether eliminated paper money and coins. this same process. Watson’s prophecy has today become a reality. office or filling station will do two things: insert an identification into the terminal located there.3 for components of EFT. many corporations deposit weekly paycheques by EFT and preauthorize account debts such as monthly interest payments and insurance payments. or one green dollar bill’.) The ACH then transfers the amount to banks where the intended recipient has an account. International systems like CHIPS (Clearing House Interbank Payments System) and SWIFT (Society for Worldwide Interbank Financial Telecommunications) have provided EFT between banks for the past twenty years. institution) is converted into electronic impulses and transmitted through a telecommunications channel to the nearest automated clearing house (ACH).3 Components of an EFT system in a store. The reasons for this are described below. As Watson predicted. Float When payments are made by cheque there is usually a two. a computerized version of the traditional cheque clearing house. repeated thousands. (The cheque itself. that is the paper on which the information is written. cheque writer.

whereas a service fee is charged for EFT. the loss of float might be balanced by early receipts. EFT will not win public support until human needs are addressed. The conversion of liquid holdings into ‘near-money’ assets such as Treasury bills and commercial paper would increase profits. Loss of float is a principal reason why cash managers oppose EFT. much touted. On the other hand. short-term loan to the cheque writer. in effect. as in ATM transactions. Cost Many banks do not make charges to people who write cheques provided that they maintain a given balance in their cheque accounts. drinking habits. Preference for cash Many people like the feel of cash. EFT spin-offs Point-of-sale terminals linked to banks and credit agencies by telecommunications is an electronic application in retailing that is technologically feasible but not yet widespread. the danger exits that stolen cards will be used to make fradulent fund transfers. The vulnerability of financial data during transmission can be protected by coding and security measures like packet switching described in Chapter 5. debts. Many people. In may be because the cheque system works well and cheque processing is low in cost that bankers appear more interested in maintaining the status quo than in promoting EFT and building a workable low-cost ACH mechanism. bank customers are given no financial incentive to switch from cheques to direct deposits. if all expenditures and receipts were EFT transactions. but the potential loss per incident is much greater. The public is not yet willing to eliminate the paper backup of cheques. In addition. In fact. the fast processing of transfers and exact information on the status of funds might enable firms to reduce demand deposits and mobilize idle money. more reliable 192 Point-of-sale terminals Electronic point-of-sale (POS) terminals that record sales transactions are found in many retail establishments. hand forms and fingerprint recognition systems. and so on. Acceptance of EFT may have to wait until the public is assured that the privacy of individuals who use EFT is safeguarded. it gives them a sense of satisfaction. Lack of public confidence Most people have had personal experience of computer errors and are wary of computer reliability. is too costly for widespread use at present. and systems security based on voice. are likewise spin-off applications of EFT that have had only limited success to date. health status. object to these types of system on moral and political grounds. Electronic shopping and home banking. Perhaps user-friendly features can be added to make the use of a terminal for financial transactions outweigh the pleasure of handling cash. Another concern of many critics is that EFT might lead to an erosion of privacy rights. That is to say. When identification cards are required for EFT access. The basic problem is lack of high transaction volume necessary to pay for these services. Most transmit sales data to a store computer which processes the data to produce accounting and inventory reports . Fewer incidents of attempted theft are reported from EFT than from conventional cheque systems. The problem arises because EFT records contain data on spending patterns and an analysis of these records can reveal personal data such as travel movements. The amount of money in transit but not yet collected is called float. too. Instantaneous EFT payments eliminate float and the possibility of earning interest on these short-term loans. the ease with which financial records can be accessed by EFT raises concerns about the security of financial data processed by EFT. a company that pays bills by EFT but receives payment by conventional cheques stands to lose money. In addition. Password codes can be breached by determined thieves. Float is. Although efficient.Telecommunications and networks time the cheque is written and the time that it is cashed. an interest-free. We next explore the promise of these systems and reasons that public reception has been poor. than passwords. The transition to a chequeless society is unlikely until both banks and their customers save money with EFT.

is an integrated circuit charge card. Home banking is not dead. Merchants will have to index their goods and provide these indexes to information utilities. Home banking For a fee some banks provide home banking for people with a personal computer and modem. listings in commercial and private user telephone directories. The purpose of this marking approach would be to identify groups that might be responsive to advertising for particular products so that advertising campaigns can be directed to them. Less common is the use of POS terminals to monitor charge accounts with credit limits within a given store. Home shopping With EFT. In the future it may become possible to key descriptions of wanted items on a home terminal and let the computer search for stores with the items for sale. the terminals are part of a telecommunications network linking them with banks and credit agencies. Marketers will then call a large test sample from the list in order to create buyer categories. banking and credit card industries. payment is authorized on the terminal by EFT. This use of EFT requires a telecommunications link from the store to the customer’s bank and the ability of that bank and the store’s bank to handle electronic financial transactions. and then try to determine ‘buying windows’ or eligibility factors that are required for a good buyer/product match using computer software to evaluate current customer-need profiles against benefit elements of available products and services. are reluctant to give up the float. licence and association lists. Most people who own PCs lack modems: to purchase one costs from £80 to £240 an investment that most people do not seem to think is worth the cost. Many banks are shifting their electronic banking strategies to small business customers and offering financial planning software for added value. In this case. the idea will be profitable. 193 . The problem appears to be that individuals. Smart cards An innovative use of information technology that will affect the computer.Messaging and related applications for management. A good idea. shop assistants will be displaced. Before teleshopping becomes more widespread. called a smart card. and buy or sell securities from home. mailing lists. home shopping has not been well received and many videotex information services that featured home shopping have failed. but home banking services have few subscribers. retailing. census and other data to obtain a master customer list. The sale of merchandise recorded on the POS terminal debits the customer’s credit limit by the amount of the sale. Although the pleasure of window shopping and the ability to handle merchandise is lost. there will probably be a surge in electronic marketing. like their corporate counterparts. Special computer programs will merge computer tapes of credit bureaux. Shopping centres will lose customers. marketing patterns will change drastically. however. open new accounts. perhaps. but they can review bank statements. Some POS systems immediately debit the amount of purchase from the customer’s bank account and deposit the money in the store’s account. Other banks are forming consortia and joint ventures to help pay for the cost of developing and marketing home banking. POS terminals can also be used for cheque and credit card authorizations. Not only can bank customers access account information and pay bills through interaction with the bank’s computer. As with home banking. another EFT spin-off. and retail space will be replaced by warehouses that transact business electronically. which writes and mails its own cheque to the payee after payment has been authorized through home banking. transfer funds between accounts. If such home shopping systems ever take hold. among others. many home shopping systems allow viewers to rotate items displayed on the screen or view them in close-up. Besides. however. home banking does not eliminate paper cheques: it merely shifts cheque writing to the bank. Selection of goods is through catalogues or newspaper advertisements called to the computer screen. But entrepreneurs seem confident that once the technology improves and a broad customer base is established which will lower costs. The challenge to home banking suppliers is to demonstrate that the services they provide save the customer money and are more convenient to use than traditional cheque accounts. home shopping (also called teleshopping) becomes feasible.

(An alternative technology is a laser card which stores credit information as tiny black dots burned onto the card surface. has begun testing the cards and talked of a five-year changeover from magnetic strip credit cards. (In these applications the cards represent electronic money. Smart cards are not a new concept a key patent on the idea was granted in 1974 when the semiconductor technology was barely two years old. they are used to pay for calls from public telephones and to pay highway tolls. health insurance information. to pay for purchases in campus shops and restaurants. protection and flexibility. at Loughborough University in England.) With their ability to store digitized fingerprints. photo and voice prints. But the idea. Visa. Bad debts should shrink because the chip on the card keeps a running tab on purchases and will not approve those that exceed the holder’s credit authorization. It could also help decision-making within a firm or corporation by making the decision-making process on-line whilst sharing enterprise resources of computing . the card will record the date and value of the transaction and a new balance will be calculated by the microchip. customers. (Unified cards add battery power.) Another use of smart cards is as follows. smart cards have attracted commercial interest for application in building security. a certain amount of money is transferred to the card’s account according to an agreement reached between the card holder and his or her bank. Cooperative processing In our discussion of the client server system in Chapter 10 we saw that it was originally designed to facilitate communications between end-user (client) and the computer processor (server). is waiting to see whether smart cards will save enough money to justify investment in them: the cost of a smart card is more than three times that of an ordinary credit card. appointments or other records). most contain an embedded microcomputer chip and permanent memory that does not lose its information when the power is shut off. such as the card-holder’s credit limit which is stored in the card’s memory. At the beginning of each month. and by the US Army to replace dog tags. the card’s microchip is not damaged. At that point. However. Midland Bank is using a contactless card that transmits signals to the card reader so that no direct contact is needed. a two-line display screen and a keyboard device to a smart card so that information can be entered or read from the card without a computer terminal.) When a purchase is made. But how about communications between end-users using the same database(s) and server(s)? This configuration could improve processing horizontally between colleagues. 16 or 32 kilobytes) which makes them able to store much more financial information than magnetic strip credit cards (e. The card is purchased for a given amount and each transaction is recorded on the card and debited until the given amount is spent and the card becomes invalid. the sales assistant inserts the smart card into a card-reader module that is tied to a host computer which decides whether the purchase should be authorized based on account parameters. like home banking and home shopping.g. saving retailers the time and cost of telephoning for credit authorization. An advantage of smart cards is their memory size (8. In case of accidents. which states that the big selling points for the cards are convenience. market pressure may force other financial service companies to issue smart cards. the amount is automatically added to the previous balance recorded in the card’s memory. if MasterCard starts mass distribution of the cards as planned. Although a sizeable investment would be required to install the readers necessary for widespread retail use. MasterCard. of course. Smart cards are used at universities to store student’s records and timetables. industry experts claim that the cost would be defrayed by reduced credit card fraud (smart cards are difficult and costly to duplicate). When paying for goods. telephone numbers. use of smart cards will gain momentum. The health industry is promoting the use of the cards for medical records which will carry warnings 194 about medications.Telecommunications and networks Smart cards resemble credit cards but instead of having a magnetic strip on the back with credit information about the card holder. allergies and chronic illness of the card holder plus name and telephone number of the family doctor. In France. Besides their capacity and versatility for information storage. such cards might save lives provided. which will in turn force businesses which rely on credit card payments to install the card readers. The first time the card is inserted in a reader each month. smart cards are favoured for financial transactions because they authenticate transactions. has been slow to catch on. more cautious. suppliers and other relevant personnel in organizations.

but it can reach a much larger audience responding to specific questions.2 Cooperative processing compared with traditional versus client server processing Traditional processing Tasking: Architecture: Software: Applications: Control: Security and integrity Path Time and place Single Open Software on client and server are independent Distributed Resides with workstation A problem because of the distributed control Hierarchical Same time and same place. can be spontaneous. video-conferencing Cooperative processing Multiple tasking with human interaction Closed Software on client and server are integrated Host-based and enterprise-wide Resides with host or server computer Has better integrity and security control because it is centralized Many cross-currents leading to problems of cooperation and collaboration Same time and different places. which. Different times and different places 195 . but it also has organizational implications that can be far-reaching: it can affect the structure of traditional decisionmaking and lead to adhocracy. unlike video-teleconferencing. Cooperative processing will not be the best substitute for interpersonal relations of face-toface meetings and the telephone. thus facilitating concentration on other productive tasks. which enables the scheduling of meetings and travel. ž Group document handling.Messaging and related applications as well as data/knowledge-bases. Groupware binds the separate activities of end-users and decisionmakers into an ensemble of cooperative processing. taking place any time and at any place. These tools include: ž Group calendering and scheduling. Such processing involves multitasking instead of the single-task processing. face-to-face meetings. ž Group meeting and teleconferencing. The processing in this environment is facilitated by software known as groupware and the processing known as cooperative processing. or group scheduling. Personal productivity programs such as e-mail. i. spreadsheets and word processing can still be used along with a DSS and EIS but integrated with group work seamlessly and without any special commands and procedures. It can overcome the barriers of time and place and offer continuous communication at one’s need and convenience. i. or group writing. It retains the advantages of distributed processing and downsizing. This enables some eye-to-eye contact and instantaneous reactions. Cooperative processing captures the flow of ideas (and discussions in teleconferencing) and creates a continuous database and knowledge-base of all information relevant to decision-making and problem-solving.2 Message handling systems (MHS) There are other message handling approaches like the Bulletin Board Systems (BBS) and the Table 17.e. Same time and different places.e. Cooperative processing has many advantages discussed and implied above. ž Group decision-making and problem-solving support through downloading of some decision-making from the boardroom to the desktop. including groupwork software utilities and development tools. Such processing is called collaborative processing or groupwork. Groupware is possible because of the many electronic tools used on a LAN that may or may not use a server but facilitate horizontal communications. Cooperative processing compared to traditional processing is summarized in Table 17. but facilitates interaction among end-users in addition to the dissemination and routing of the necessary shared data/knowledge from groupware servers as well as application and file servers. It utilizes client server computing. More intensive use of e-mail messaging.

There are four main approaches to the integration of message handling. synthesizes and communicates information.) offered by Information Service Provides and will be discussed in Chapter 19. shipping and receiving of goods. This was the first generation of message handling that lasted till mid-1980s. The most popular and successful in the mid-1990s was Software Notes by the firm that produced Lotus. e-mail. These processes included price generation. The application of information .Telecommunications and networks Newsgroups. EDI is very cost-effective and has a fast response time for applications where it can be used. The third generation of technology saw video-conferencing. All the vendors incorporate multimedia processing as part of the broadband approach to information transfer. purchase ordering. Then the second generation that lasted till 1990 saw teleconferencing. Despite the commonalties between the messaging vendors. Also. Microsoft had a back-end server and front-end client with an integrating platform for enterprise-level communication. etc. This was in direct competition with the other contender in this field. Whether the creative programmers and management responsible for the success of Notes will find the large umbrella of IBM and miss their informal and ‘small’ environment of Lotus will not be known for years. In this concluding section of this chapter we are concerned with the integration of message handling systems that include e-mail. there is resistance to EFT which can be traced to the fact that people like to have cash literally ‘in hand’. there are some differences in the architecture of the systems. 196 Summary and conclusions Message handling has seen an evolution in technology and organization. EFT is concerned with the transfer of funds once the paperwork has been satisfactorily completed by EDI.3. in 1995. The public sector technology consisted of the telephone and the mail. All these approaches are basically extending messaging into discussion databases.4. invoice processing. This leaves two other main messaging vendors in the field: Microsoft with its Exchange system and Novell with its Collaborative Computing Environment. The accelerated income velocity of money under EFT has a side-effect. Electronic messaging is thus fast becoming an infrastructure providing much of the services that the traditional operating system provided. product availability reporting. Technologically. accounts receivables and payable. Lotus had a backbone for routers. and a tight integration of its e-mail program and Notes. One advantage of EFT is that a large number of transactions involving large sums of money can be processed quickly and efficiently. IBM made a hostile and successful takeover of Lotus. 1995). and responses to FQAs (Frequently Asked Questions). the integrated massage handling system. EFT and EDI. group scheduling and electronic forms within a overall and overreaching strategy for its deployment throughout the enterprise. electronic shopping. This evolution is summarized in Figure 17. EDI was appropriate for only highly structured and well defined processes and hence did not include many of the processes in a typical office and business. These are often services (along with e-mail. Many vendors have acknowledged the corporate needs for downsizing and using the client server platform for the enterprise left the mainframes behind. However. IBM had a middleware platform with a multiprotocol backbone for scheduling. Not all groups of potential end-users have the same reaction to EFT because they are affected by the advantages and limitations in different ways as summarized in Table 17. Newsgroups and workgroups of collaborative processing. Then. EDI can be characterized as an application-toapplication or computer-to-computer processing using teleprocessing and networks that automated many an office and business process. There are also problems with standards especially for the transmission of EDI documents. workflow automation and other message-based applications. It is one of the technologies that manages. Corporate mangers must contend with the loss of float and the possibility that security and privacy of financial transactions may be breached. It could not handle problems that needed a human dialogue and is not considered a very end-user friendly system for clients and suppliers. And Novell offered cross-computability with any server through the application components provided by its groupware (Bragen. cooperative processing and the developing MHS. BBS. Despite these difficulties. an interenterprise connectivity. The earliest messaging systems were centralized both in the private and public sector. IBM. EFT and EDI may lead us to a cashless society.

1989. especially when the transactions are between individuals and businesses as is the desire on networks such as the Internet. leased from a common carrier and can carry both voice and digital messages at 1. The firm will be able to use a few dozen internal video-conferencing rooms to hold meetings with 30 to 40 customers and suppliers around the world.3 For and against EFT For Corporations: Better cash management Security problems Retailers: Quick credit approval and transfer Financial: Lowers transaction institutions Fear of monopoly Consumers: Direct payments/ receipts Home transactions now possible Against Loss of float Case 17.54 megabits per second. New Jersey (US). . Source: Information Week. as the computer system driving the video-teleconferencing. The cost of EFT and EDI is also a drawback. providing seven-digit dialling for GE offices world-wide. The replacement of plastic cards with smart cards may help reduce the use of cash and cheques if they can be secured. We will return to this topic in Chapter 20 on the Internet.4 Evolution in message handling Table 17. London and Paris with the Rembrandt series of Codes by Compression Labs Inc. (A T1 link is a telephone line. the network will link 20 offices with offices in 25 other countries.) Teleconferencing will help GE customize its products for specific markets in different countries and implement new production and engineering techniques. 197 Capital costs Costs of ATM Loss of float Invasion of privacy possible technology and telecommunications to home banking and home shopping is moving slower than many had expected. will comprise many T-1 links and three intelligent nodes at Princeton.1: GE bases global network on teleconferencing General Electric Corporation (GE) has launched a two phase. POS (Point of Sale) terminals and the consolidation of backoffice automation in banks have high priority and are consuming the scarce technical resources.Messaging and related applications Third Generation 1990− MHS Cooperative processing Video-conferencing Second Generation 1984−90 EFT EDI E-MAIL Teleconferencing First Generation 1978−83 Centralized Private Public Telephones Mail/Post Figure 17. Implemented by AT&T.. June 5. The network which will be managed by GE’s network-control centre in Princeton. One reason is that ATMs (Automated Teller Machines). three-year plan to use videoteleconferencing to promote information sharing between its US operations and its international joint-venture businesses. It hopes this technology will help the firm solve day-to-day problems as if the factory were in the same building rather than thousands of miles away. France Telecom and British Telecom.

H. Dec. and customer declarations. Group dynamics. (1992). R. less than 1% of all business transactions in the UK are via EDI. Bragen. 14(8).. (The latter coordinate EDI networks. July 19. and Rein. p. 13. T. Video teleconferencing: the state of the art. Two factors contribute to Britain’s lead over other European nations in EDI activity: the liberalization of British telecommunications and the ease with which value-added network suppliers can enter the market. D. Source: Information Week. The Internet and EDI.E. Telecommunications. International videoconferencing: a user survey. Ninke. 8. 1994. Bibliography Aldermeshain. Ellis. 287 293. K. 103 111. J. R. 72(1). cooperative processing. P. Client/server vs. (1990). Trauth.M. customs and Home Office. and Pile. D.M. (1993). Global Information Management. Four paths to messaging. Information Systems Management. G. invoices. The Internet Magazine. EDI appears to be a major facet of the proposed Police National Network. 121. S. (1994). 5 15. M. and Cukor. (1991). (1992). July. J. R. S. Electronic data interchange: a new frontier for global standards policy. Thuston.A. Muiznieks. 221 236. Issue 10. Journal of Information Technology. (1994). Gordon. 14(8). (1995). O’Keefe. Desktop videoconferencing: candid camera. pp. (1992). 2 6. 71 136.J. and Wilkes. (1993). Riaczak. 29(11). They use networked computer-based systems to process orders. Special issue on ‘Organizational Perspectives on Collaborative Processing’. Supplement 17. V. 26(4). Computers and Secusecurity rity. Do the business. P. McNair. p. 9(1). F. 2. 7 14.A. (1993). June. reduce errors and minimize the amount of paperwork their offices handle. Data Communications. Journal of Global Information Management. D’Angel. AT&T Technical Journal. W. Security in electronic messaging systems.1: Costs of message handling and related processing 1974 Telex Fax Electronic messaging Processing (MIPS) $90 Storage (gigabytes) $10 Sending 1 megabyte of data across the US $0. 1989. (1993). E. July 1995. (1992). PC Magazine. Cost of EDI in 1994 (with an Internet service provider Start-up costs Telecom costs Access costs (based on 500 2000 documents/ month) 198 $50 $14/month $45/month . and Thomas. 9(2). Groupware: some issues and experiences. Friend. 1995. (1992).20 Source: Fortune. p. Van Nostrand Reinhold.J. 18 and Computer Weekly. 1(3). Communications of the ACM. 11(3). Slashing paperwork should help reduce the time that prisoners are held on remand before trial by speeding the time it takes to process and exchange information between the police. 1(4). 38 48. Sources: Financial Times. Wright. in order to save manpower. 6 17. Standardization of information systems technology at multinational companies. May/June. AT&T Technical Journal. Morley. Videoconferencing: the bigger (and better) picture. Senn. 33 34. Gibbs. K.Telecommunications and networks Case 17. January 63 64. Labriola. Oct.2: Electronic data exchange (EDI) in the UK Over 2000 companies in the UK are engaged in EDI. 38 58. Emmelhainz. net. B. M. Electronic data interchange: a total management Guide. 139 151. 1989. Coopock. Which Computer? 15(7). C. The video communications decade. 70. (1995).S. (1995).. H. 4(1). sometimes known as ‘paperless trading’. Telecommunications. (1995). 45 48. S. freight and forward notices. Richardson. (1995). Telecommunications.A. 7 20.L. Electronic data interchange. Telecommunications.) Nevertheless.95 1994 $25C $10 20 $1. 19 30. 119. PC Magazine. Data interchange and legal signature surrogates.70 $20C $3 $0. 64 66. Information Systems Journal. Global messaging: the role of public service providers.A.

video-conferencing.1. animation and full feature films. In some PCs. teleshopping. 1989 Have nothing in your house that you do not know to be useful or believe to be beautiful. Spatial composition links various multimedia objects 199 . touch. But to appreciate multimedia applications we need to understand the nature of multimedia. telebanking and financial applications. A list of current and potential applications appears as Figure 18. Then there are applications that are not feasible without telecommunications such as video-on-demand. multimedia comes as a standard capability. and the resources needed for these applications. Multimedia and distributed multimedia Multimedia is an extension of the traditional media of numeric data and text. The extension of multimedia processing is when it is available from a distance through networks and telecommunications. keyboard. William Morris Introduction Multimedia is nothing very new conceptually speaking. Multimedia not only adds to knowledge but also to the friendliness of the computer interface. Perhaps we will get used to the novelty of voice messages but then there will be other multimedia approaches to make you feel welcome and happy working with a computer. But multimedia in these cases have only enhanced existing applications that existed without telecommunications. mice and other devices. joysticks. are what is often referred to as multimedia. But all these devices do not necessarily require telecommunications. Multimedia systems include multiple sources of various media either spatially or temporally to create composite multimedia documents. Later there were graphics added on but that did not have the resolution and quality of pictures. Then there are specialized devices with the ability to compose music and do computer-aided learning (CAL). email that delivers multimedia documents. there are many applications of multimedia. Then we have access not just to the local library of multimedia but to all the libraries strewn across the country and indeed around the world. however. its characteristics that are relevant to computer processing. a couch potato medium. and interactive TV. and games. This is where we start this chapter. distance learning and telemedicine. the digital library. The addition of graphics. sound and touch (of the keyboard). cooperative processing. with most workstations. images and pictures. Some applications do use telecommunications like telecommuting (using computers for work at home). George Gilder in Microcosm. telenews. as well as audio. television is inherently passive. It is so much easier to communicate with all the senses of sight.18 MULTIMEDIA WITH TELECOMMUNICATIONS In an age when computers will be voice. The sound of one’s name and a welcome message or the sound of ‘YOU Have Mail’ is often much more refreshing than just seeing the message on the screen. It is these latter applications that we will examine in this chapter.

. Thus isochronous processing is similar to point-to-point circuits rather than packet switching where synchronization is not crucial in say e-mail delivery where the processing is actually not isochronous and data is moved in packets around possibly different routes at different times with small delays hardly perceptible. large capacities are required for the transfer of the multimedia messages which means more bandwidth. communicate and even play. and even better diagnostics by a doctor who is otherwise not available on-hand. which is best appreciated by looking at Figure 18. . . Temporal composition creates a multimedia presentation by arranging the multimedia objects according to temporal relationship . . That is why packet switching is feasible for many text-based and applications-based information including e-mail say on the Internet that uses the TCP/IP protocol. 1994: p. To better 200 understand these demands and needs it is necessary to look at the basic characteristics of multimedia transport which creates these needs. The reduction of delay is important in realtime applications such as in video or the showing of a film. One is that more storage space is required at the server and client receiver end. Networks . This has two consequences. It will contribute to the way we work. This enables richer video-conferencing. not only by increasing the range and variety. multimedia could help in our work by improving communications and increasing dissemination of information and knowledge.2. learn. These are filtered and then available for browsing and sometimes even manipulation. The second basic and important characteristic of multimedia is that it is very memory intensive. dealing with object size. (Furht. Requirements of multimedia The distribution of multimedia makes many demands on computing resources. It could improve one’s enjoyment of films and video. One basic need is the timing of the multimedia components and the need for low latency. Doing this using networks enables one to do collaborative processing using this vast and rich base of data and knowledge in all its many manifestations in different media. more meaningful learning and training even from a distance. rotation. Multimedia can also be viewed as a set of objectives and data entities which are a multidimensional array of numbers derived from various sensors that record images.Telecommunications and networks Music composition Computer-Aided Learning (CAL) STANDALONE Interactive TV Home PC / Workstation Telecommuting Compound document e-mail ENHANCED Teleshopping Telenews Telebanking/Financial transactions Collaborative processing Games On-Demand Digital Library Video-conferencing Distance education Telemedicine Video Films Interactive Passive DISTRIBUTED Figure 18. and secondly. and the placement within the entity.1 Applications of multimedia into a single entity . When interactive. 53). Would you not be unhappy if the sounds were not precisely synchronized with the corresponding movements of the lips? This synchronization is referred to isochronous processing and refers to real-time communication that ensures minimum delay by synchronizing to a real-time clock and establishing a virtual channel. Say that you are seeing a closeup of a singer. but by making them available whenever and wherever they were wanted. audio and videos.

storage is a problem and then one must resort to compression. Multimedia. There are many techniques of overcoming the memory intensive problem. in image processing of a film it is not necessary to capture and store all consecutive image stills in sequence but instead to capture and store a base image and then only the changes to that base making the computer processor do the construction of each changed image.0 6. Table 18. retrieved and manipulated in large quantities and at high speeds. . The need for large bandwidths. Different media have different needs of bandwidth depending on the nature of the application as shown in Figure 18.2 Latency defined designed and optimized for data and textual traffic are not suited for multimedia traffic like video and films. However. (1).1. The rise of enterprise computing. Compression over the years has greatly improved as shown in Table 18. These resources have to be managed along all the paths between the end-user at a client device and the server. the 201 A factor of 75 over 24 years Source: Adapted from Joseph Braue (1992).4 100 500 550 1000 Source: Adapted from Borko Furht (1995).2 where different types of multimedia are compared. 21(12). Data Communications. transported. video and audio stream playback and teleconferencing are both real-time applications but with different latency and delivery requirements with teleconferencing requiring very small latencies but playback applications requiring guaranteed delivery of real-time messages.Multimedia with telecommunications Access time Latency Transfer time Instant at which delivery is completed Instant at which control unit initiates call Figure 18. storage can still be a serious problem as demonstrated in Table 18. SMDS (Switched Multimegabit Data Service) and SONET (Synchronous Optical Network) discussed in earlier chapters. For example. 48. multimedia must be efficiently stored. Resources needed for multimedia processing Large bandwidth needs of multimedia require switching capacities for large megabyte and gigabyte transfers and will need equipment like ATM (Asynchronous Transfer Mode).3.2 Storage requirement for typical applications (in megabytes) Text of 500 pages (normal standard) 100 Fax line images uncompressed 10 minute animation (15 : 1 compression) 100 colour images (15 : 1 compression) 10 minute digitized video (30 : 1 compression) 1 hour digital video (200 : 1 compression) 1. Even then.1 Progress in compression over the years for data rate acceptable for video Year 1978 1980 1982 1984 1986 1988 1990 1992 Bandwidth (Bits/s) 6000000 3000000 1500000 760000 224000 112000 112000 80000 Table 18. despite compression and high compression ratios. Multimedia systems: an overview. 1. 27. Once compressed. One is the use of techniques for economizing in memory. For example. low latencies and real-time delivery makes many demands of computing resources.

and the client. Now multiply the storage for one delivery with the large number that will soon be in demand. Servers The servers are specialized processors designed for handling large masses of data.3 Demands for bandwidth. An overview of the main components of a multimedia system is shown in Figure 18. Not to exact scale HARDWARE Multimedia Storage A U D I O Network V I D E O Multimedia Server Figure 18. They will vary with the type of multimedia being processed but whatever the media the server has to be fast 202 and powerful.Telecommunications and networks Bandwidth 100 Mb/s Visualization 10 Mb/s Desktop video-conferencing 1 Mb/s 100 kb/s Integrated video-conferencing Digital videophone 10 k b/s Analogue phone Time Figure 18. It must also be supported by large and often very large storage capacity. We will discuss these in turn starting with the server end. These storage devices must be fast enough to serve the high volume of multimedia without being a bottleneck in processing.4. servers must also be equipped with multiple disk drives with video files for each . The magnitude of storage space required for multimedia has been mentioned before.4 Multimedia via telecommunications server itself. For simultaneous access. Even a modest choice of films and video can quickly run storage into gigabytes.

. text from a word-processing program. 41).. Networks We have mentioned the need for bandwidth earlier. Besides bandwidth.5. there is a problem of architecture with multimedia. .5 ISO vs. or a fully fledged film or video. represents the logical storage of multiple forms of data contained in the multimedia information . 1994: pp. By providing an organized method for computers to logically communicate. Intelligent hubs. 42. Microsegmentation. which is fast becoming the industry standard for large PCs and workstations. does rely on better switching as well as its integration with bridging and routing especially if a hub is being used. The node layer. . Nowadays. Multimedia needs a new applications layer to accommodate multimedia applications and the transport protocol needs an additional control capability to prevent ISO Application Presentation Session Clients The client processor has to be large and powerful though not as powerful as the servers. The objects could then be data from a spreadsheet. One approach to the problem of bandwidth is microsegmentation where users are assigned to segments of the LAN. the Pentium is recommend for any respectable performance for multimedia processing. Multimedia needs the addition of a base layer and a node layer. residing over the transport layer. the transport protocol needs the means to package timing traffic with audio and video to prepare them for transmission. (Domet III et al. congestion at the destination and at the routers. however. the layer model ensures data can be transferred independent of systems platform . This reduces the users per segment and better serves the users in terms of bandwidth available to them. . Objects are also embedded in applications but can be retrieved by OLE. it is hoped. Storage can be done by storing multimedia as objects in the object orientation methodology.Multimedia with telecommunications drive. Besides the CPU requirement there is the MULTIMEDIA Application Presentation Session Base Transport Node Transport Network Link Physical Network Link Physical Figure 18. The interoperability ensures that different platforms can equally access multimedia information distributed over networks regardless of the incompatabilities in their operation. will be able to do this and provide bandwidth-on-demand as needed and provide virtual switched circuits. Also. multimedia architecture 203 . Early clients were the 386 or 486 processor. It corresponds to the OSI layer architecture discussed earlier and now compared in Figure 18. Object Linking and Embedding methodology from Microsoft. The base layer is added to the OSI model in order to coordinate the data access and make the data generated by these queries systems independent. Most upgrades to the transport and other layers required by multimedia can be handled by software upgrades.

voice and video that can be sent in one form and read in another. full user interactivity and personalized information channels.Telecommunications and networks need for the audio and visual coder and decoder. that has interactive program guides. It is therefore safe to assume that implementation will come in stages and will come first in the office rather than the home because the offices will be able to afford it and make good use of it. also called a home communication terminal. It defines display formats with specific resolutions and provides processes for reading. Some systems come with a decoding board or a coder/decoder called the codec which needs to be equipped to both the client and the server. For example. H. Here we have a standard before the technology has stabilized. Also. There is some uncertainty about how many participants there will be and how much each will participate. H. video information and the organization into blocks and groups of blocks. movie-on-demand capability. H.243 defines exchanging a unique encryption key at start of compression. Sending video mail in the store-andforward mode does not require isochronous processing but video mail does demand minimum delays and consume prodigious bytes of storage.320. Applications of distributed multimedia Even without discussing the more exotic multimedia application of telemedicine and digital libraries. There are also special standards being developed for codecs such as the H. MPEG will test the rule of which comes first: the chicken or the egg. The MPEG standard is the best known since it is an international standard approved by the ISO though it does not carry the ISO prefix. 204 Video-conferencing Another important application of multimedia in the office is video-conferencing. Standards Multimedia needs its standards and there are plenty of them. But even then there will be a high cost since these applications will be very storage-intensive. Other combinations of e-mail are also storage (and bandwidth) intensive like text augmented by audio. e-mail and retrieval of multimedia information for decision-making and problem-solving. It is a two-way interactive communication with both ways possibly containing multimedia unlike the downloading of a film or video which is a one-way downloading.242 and the umbrella standard of H. There are other standards like H. This gives a quality comparable to the videos on rent in many stores assuming a 150 Kbyte/ps bandwidth. Examples are telecommuting. Eventually we will have compound document email with text. H. Some codecs achieve 175 000 pixels/frame which comes close to high quality TV which is between 190 000 200 000 pixels/second. The most prominent is the MPEG series. where MPEG is the acronym of the organization that developed the standards: Motion Picture Experts Group. we can see that the implementation of distributed multimedia is going to be neither easy nor cheap. The MPEG-1 standard defines a video resolution of 352-by-240 at 30 frames per second.221. an e-mail containing a one-minute fullmotion video will range between 10 Mbytes for VHS quality. It is already possible to send an image document and review it as text (using an optical character reader) and have text read aloud (using voice synthesis). the standards or the technology. These characteristics are: 1. robust graphics. and images with audio. . The MPEG is being watched by the profession because it is an exception to the rule that you need a stable technology before a standard is defined. Many codecs also perform compression. at least seven in the US alone. The codec is the multimedia equivalent of the modem (modulator/demodulator) we visited earlier. There are other standards organizations involved in developing multimedia standards. The client could also have a set-top box. 2.230. It has bandwidth needs that are on-demand but an unknown size demand unlike say the downloading of a known sized full-feature film or video. The widely accepted standard may encourage the industry to bring out new products but it might also inhibit innovations and freeze the technology of multimedia. MPEG deals with moving images while JPEG standards deal with still images.261. decoding. text with images.261 by CCITT. It may be considered an extension of existing teleconferencing but it has many important characteristics that make it complex to implement. implementation will be first in the areas where existing applications do not yet have multimedia and need it.

network and transport and is then decompressed before appearing on the screen as the desired image with its multimedia package. network and link to the network. improvements in size of image.221. engineering drawings. multiple call set-up. H.320. video and images in transmission. dynamic allocation of network resources. there are some distinct directions that video-conferencing is taking. text and other relevant multimedia information. On reaching its destination client. it flows through the layers of link. the ability of participants of videoconferencing to take notes. SERVER STATION Multimedia conferencing utilizes the concept of shared virtual workspace which describes the display at each client station and where each participant can send or receive data. voice or video as part of collaborative processing. Other implementation problems concern hardware interrupts and interoperability problems and the global adoption of international standards on video-conferencing quality despite the existence of many standards for video-conferencing like the H standards from the CCITT concerning interoperability: H. and graceful degradation when faults occur (Furht: 1994: p. USER STATION Grab frame and encode Compress Decompress Communication transport and network layer Network Communication transport and network layer Link and Network Adapter Link and Network Adapter Figure 18. and then communicated through the layers of transport. Furthermore. quality. The basic process of video-conferencing is shown in Figure 18. For both cases we do not show all the layers nor all the components involved and stick to the basics to keep things simple. and the development of a library of all relevant multimedia objects that are easily and quickly retrievable. compressed. change memos and diagrams.242 and the umbrella standard H. all during the video-conferencing. 58). picture and colour support.Multimedia with telecommunications 3. One of the unresolved problems of multimedia conferencing is the optimum communication architecture for composite streams of audio. What we need is a high resolution video and audio overlayered with spreadsheets. The process for the opposite direction is the process in reverse. synchronization of shared workspace. It is real-time and hence isochronous processing.261. the functions performed include real-time control of audio and video. It will include the upgrading of telephones and the downgrading of meeting rooms leading to desk video-conferencing. and manipulate data on spreadsheets.6 Video transmission 205 . the initial use of the PC with live video in a window of the screen. Despite these difficulties. H.230. jot annotations.6 where the object to be transmitted over the networks is found stored at a server. H.

telemedicine was first supported by the US military) or for getting medical attention where it is not possible such as in rural areas or when travelling away from the family doctor. The impact of telemedicine is on both the patients and the doctors. they also want to control the distribution and so there is a confrontation between the holders of content and the distributors which is resolved in the market-place through mergers. This may require a lot of sophisticated equipment for data acquisition at the patient’s end but it may be the report of a nurse on a helicopter escorting a wounded person in a car crash (or in a war zone). Some operations have been performed remotely by a surgeon on a dummy with all the movements being transmitted electronically to a robot working on the patient. Before we examine these applications. nose and throat. treatment may include an operation. Telemedicine Telemedicine is the delivery of medical care to a patient by an expert (or institution like a hospital) separated by distance but connected by network and a multimedia communication system. blood pressure. the educational sector. A package of information on the patient is multimedia in that it may contain CAT or MRIscans and other computer generated pictures. Conceptually. We shall discuss these relationships later but only after we identify another group of players. for they also generate and own content. However. etc. They become important players in the market for they control the content. The main technological differences lie in that the multimedia in video/film-on-demand is transported only in one direction and that the storage at the server end has to be very large in capacity and very fast in its retrieval. collection of data from special instruments. And what if the patient lives in a remote area of the . Also. and the past medical records of the patient. all under the control of the end-user while in the relaxed comfort of the library or bedroom at home. We have already discussed the storage considerations but it is important to note that what is stored is mostly entertainment and that brings the entertainment industry into the picture. Thus telemedicine is not as complicated as video-conferencing in that multimedia is often not exchanged but transported one way. This raises many important issues in telecommunications and networking along with moral and ethical issues of who has the priority in traffic congestion. In some cases. All this is sent by telecommunications using a wideband telecommunications system capable of handling multimedia traffic. One impact will be on the patient who will no longer have to go to a family doctor but a medical facility equipped for telemedicine and have a diagnosis made by a doctor or a specialist. testing of blood samples. zoomed for any frame. and a verbal opinion or set of questions by the patient’s doctor to the expert.. The collection of data would include the recording of data such as pulse. we look at another application which is somewhat similar to video-on-demand. run in slow motion or at fast speed. The great advantage that both video and film-on-demand have is that they can be viewed. stopped and replayed. This group includes the digital libraries and educational institutions that offer distance learning. this package is not grabbed from a server as in video/film-ondemand but is created in real-time with current pulse rate and heart beat transmitted as they occur. We will now define telemedicine as being the collection of data on a patient in digital form to be transmitted remotely to another point for analysis or diagnosis. alliances or takeovers. the observation of physical conditions like colour of the eyes or face through imaging devices. rolled-back. that of telemedicine.Telecommunications and networks Video/film-on-demand This is where full-length films or videos are available on demand and delivered to the office but most likely in the home and on the home entertainment device. The multimedia 206 package may be sent only one way but the dialogue is a two-way dialogue and maybe not between many people as in a video-conferencing situation but between at least two people on an on-line real-time basis for time could be crucial in many a situation. It could be used for getting an expert opinion or any opinion because of the urgency involved such as in the case of injuries on a battlefield (incidentally. The expert looks at the multimedia package and may ask questions and a dialogue may ensue after which a diagnosis is made. observations made by special cameras that can go into orifices like the ear. entire medical case history in text and graphics. there is little difference between the processing of videoconferencing and the very common teleconferencing.

The end-user could browse through the index and even glance through the contents or the book itself on a terminal and finding what is wanted would have the article or book downloaded to the terminal. Here books and articles can be searched by content. There was much talk about the distribution of royalties and then there was a long lull. 21). The family doctor will no longer have a monopoly over the local patients but may have to start seeking remote patients. The impact on the family doctor will be access to expert opinion instantly but a loss of patients over time. Many issues are raised by telemedicine. there is some philosophical and legal argument about the relevance of intellectual property law to software protection. the telemedicine facility. In fact. One concerns the conversion of all the material in our libraries into tangible objects with electronic digital representations such that they can be easily accessed and retrieved.) And then there is a capital cost (the Biblioth`que Nationale that opened in 1995 had e cost around $2 billion). or just lives in a developing country without good medical facilities? Telemedicine may then be the only good answer available. at what location should the doctor be licensed? At the location of the patient. What happens if a document or message for one patient gets switched to another patient? Who is responsible? The doctor with the patient. Also. 1996: p. checking-out. This will change the printing and publishing of books and magazines as we now know it. envisioned one library in all of the US receiving queries on a book or article. the designer of the programming language BASIC. or the telecommunications carrier? How is the telemedicine technology to be regulated? What is the liability of the providers of the technology that makes telemedicine possible. 207 Digital library In the 1960s and 1970s there was much talk of an electronic library. Using a digital library (or the Internet). (The operating costs of the Biblioth`que Nationale de France costs approxie mately $260 million for its 5 kilometres of computer controlled belts that deliver books to some 150 points. a document can be available immediately after it is written and put in machine readable form. advances in networking made transmission of book material feasible and even affordable. things are getting worse since networking and the availability of holdings in the digital libraries may soon be accessible anywhere in the world by countries (especially in the Far East and the Pacific Region) that have not agreed to the adoption or enforcement of international agreements to protect intellectual property rights. copied and then distributed. or both? Who is responsible if there is a misunderstanding on ‘this’ or ‘that’ artery. Some in the software industry believe that ‘existing property laws are fundamentally ill-suited to software. Books in a library on some scientific and high tech subjects get outdated sometimes even before they are printed. This is an old issue and is not getting near to any solution. (Davis et al. Also. John Kemeny. Then in the late 1980s there was a rebirth of the idea the term digital libraries surfaced. The problems are rooted in the core assumptions in the law and the mismatch with what we take to be important about software’. limb or organ resulting in the wrong actions being taken and an ‘injury’ to the patient. including that of the telecommunications and network provider? not hurt. such as: How is the cost of the visit for telemedicine split between the local doctor and the remote expert? Given that medical practice has to be licensed by each country and even each state (or province or county) in each country. checking-in and reshelving of books in a library. But there are many unresolved problems and issues. One relies therefore on journals which also have a long lead-time for publication. or is remote from home. A grant from the NSF in the US in 1993 to six universities for a digital library initiative did .Multimedia with telecommunications country.. There is also the issue of the protection of intellectual property rights. the expert. The interest was also sparked by the falling prices of costs of digital storage relative to the costs of library shelf-space and the increasing costs of the labour-intensive activity of cataloguing. the transmission equipment manufacturer. But the transformation to digital was perhaps largely the result of the emergence of the ISDN (Integrated Services Digital Network) and the promise of B-ISDN with high capacity bandwidth designed for multimedia transmission.

there is a general acceptance that digital libraries will contribute to formal learning.7 Distance education 208 . Stanford. Harvard. Cultures and communities do not. Instructor Multimedia workstation Distance learning Distance learning is education in a place that is remote from the sources of learning whether this be instructors or teaching materials. why not also have lectures delivered electronically? Why can a student not learn from the best instructor and researcher in the country or the world.) either by the student directly or on the recommendation by the instructor. Special hyperlinked versions will facilitate search and use. What will be the impact of a digital library on levels of education and the acquisition of knowledge? How will it affect the delivery of educational services? Do students need to go to the library or can retrieve what they want from a room in the dormitory or from their homes? In addition to books being delivered electronically. encyclopedias. Distance education along with digital libraries if at all accepted as a viable concept will have a profound impact not only on the way we educate the next generation but also on the infrastructure needed for such changes. It will greatly support distance learning which is the subject of our next topic. This mode of education that blends networking with educational materials (including appropriate courseware with varied media providing richer insights) not only implies different modes of instruction but also different modes of learning. just as a patient get the best advice from medical experts wherever they are? Why can any student not have quick and easy access to the lectures and research findings in Yale. atlases. This is achieved by transmitting educational material like courseware designed for remote learning in multimedia packages to the point where they are needed. Oxford and Cambridge? How will this affect international education? Will the digitization of information make it less accessible to a class of information-poor who must then compete with the information-rich class? How will digitization contribute to economic development? How will this affect productivity and competitiveness (nationally and internationally)? Will it improve our life-style and increase our standard of living? Whatever the answer to all the above possible questions. In technological and social changes there is always an interplay. informal learning and professional learning and may even contribute to digital schools. between the forces of conservation and innovation. let go COURSEWARE ON VIDEO SERVER Data/ knowledge− base NETWORK Students at their multimedia stations Figure 18. or both. etc.Telecommunications and networks title or author.7. and should not. Also accessed are materials from data/knowledge-bases (which may include multimedia dictionaries. a tension. The instructor using a multimedia workstation will instruct courseware to be sent through a network to a pupil or student who also sits at a multimedia workstation to receive the material. The process of distance learning can be visualized as in Figure 18.

This could be done using a digital library or through communications media.3 Comparison of modes of delivery Telephone Main users: Content: 1/2 WAY: Bandwidth needed: Security: Usage: Access: Topology: Switching: Protocols: Billing: Everyone Low 2 way sequential Very low Good Very easy to use Excellent Bus/star Circuit switching POTS Per unit used Cable TV Homeowners Low/medium 1 way only launch. The turn-round time from a finished manuscript to the bookshop was reduced from what was then around 1 2 years to 1 2 months. A contract for a book in 1990s often had an additional clause giving the publisher the copyright for use of the material for all multimedia publications to be done electronically if deemed profitable. Publishers are thinking in terms of making all publications multimedia and even interactive: a fiction book would have an appropriate background music for each chapter. and later editions Easier access of search and retrieval by different identifiers using computer indexing and hypertext methodology The entire publishing industry will be environment friendly and save the cutting down of trees for making paper currently used in publishing. a text in chemistry would have a film or video on an experiment being performed. 1995: p. Multimedia electronic publishing Electronic publishing is the publishing of material by electronic means like a computer with a fast laser printer. the user. Partly because there is a high cost of capital and conversion and partly perhaps publishers are thinking of multimedia publishing. analogue Good facility High Not needed Very easy to use So so Trunk & branch Unswitched Proprietary Per connection Source: Adapted from Furht (1995: pp. 83).Multimedia with telecommunications lightly structures and practices in which they are invested heavily. 209 . and which innovations to reject or adopt. This explains the great interest in communication carriers and the telecommunication industry by the publishing industry and the many alliances and mergers taking place not just in the publishing and entertainment industry but with carriers and the computer industry. and an encyclopedia entry on space travel will have the film clips of the actual space Table 18. Networked (Internet) Anyone Medium/high 2 way sequential Low/Medium Not good Not so easy Not so good Routed Packet switched TCP/IP Poor facility Multimedia networked Everyone High 2 way interactive High to very high Poor/good Easy Not so good Star Switched/unswitched ISDN. (Levy and Marshall. 28 9). revisions. Integration of electronic publishing would have many advantages such as: Reduction in current prices of publishing and distribution Reduction in the time required for production and distribution Ease and speed of updates. The task in the years ahead will be to decide which existing practices and structures to let go and which to retain. It will also greatly reduce the costs of books once book publishing is integrated with distribution through telecommunications and networks. Such interactive multimedia can add greatly to learning especially to electronic distance learning. be it cable TV or by telephone to a computer screen or to a combination device. And yet electronic publishing is not very common. can get more information (or action) from any of the objects on the screen. even a child. In the early 1980s we had a special computer program TeX that electronically published books for computing with symbols used in mathematics and science. With the click of a mouse.

If its publication and its distribution can be integrated then why not have one office do all the editorials. there is the group of providers of communication services. satellite and the Internet services. There is fierce competition amongst some of these firms especially in the communications provider industry.3. which include the companies that offer telephone. and manufacturers of peripherals and supporting telecom equipment. This then could be a battle for survival. however. has a comparative advantage as shown in Table 18. implications. Most of the Organizational implications of distributed multimedia For all the many applications of distributed multimedia. software houses. if not marginalize.8 Players in the multimedia market 210 . add art work and coloured pictures. This group include manufacturers of computers. But the publishing industry includes the publishing of newspapers too. The important implication results of many large and important firms and industries in the distribution of multimedia. then can the dangers of concentration of publishing power be contained and curtailed? There is no question that telecommunications and networks make many an application now possible but along with it there arise many social and ethical questions and issues that we may not be ready to handle and cope with. compose the news stories. We mentioned earlier that the entertainment industry has become most interested in the distribution of multimedia films and videos-on-demand. and then transmit this to the regional and local points where local or regional news can be added and printed off fast electronic printers? Wouldn’t this be cheaper and more efficient? But would this also lead to a concentration of power assuming that information is power? If so.3 We have thus far discussed the publishing of books and magazines and journals. There may be organizational problems too. The industries involved in multimedia communications face each other in the marketplace as depicted in Figure 18.Telecommunications and networks Integration requires the consideration of different modes of delivery which are compared in Table 18. This includes the publishing industry with Goliaths like Simon & Schuster in the UK. Another group interested is the computer industry which generates the products used for multimedia. the other. For example. it that all that bad? If yes.8. Video Publishing Other Govt. Libraries Education Health− care Others End Users Soft− Telecom Others ware suppliers houses Computer Industry Figure 18. It is likely that all will survive but only after much restructuring and realignments. And finally. We have also mentioned the end-users implicitly as including libraries. the telephone and cable TV each think that they can produce one device that will do away with. the health-care institutions like hospitals. collect all the raw data. wireless. Each. we have examined the technological implications but not the organizational Entertainment Industry Film Telephone Cable TV Providers Internet Satellite Wire-less Hard− ware mfrs. and the government who are interested in the welfare of their citizens and subsidize some organizations like libraries and hospitals. cable TV.

to the point that one can predict (with some caution) that we will soon be able to see any full feature film or video. They will either do it through fair competition. like the home theatre. These factors vary with applications. One way is shown in Figure 18. high volume and size of transmission. This will be decided in the near and far future and so will be discussed later in the chapter entitled ‘What Lies Ahead?’. still & colour-bit images Motion video Motion JPEG MPEG-1 50 MHz 486 240 MB disk OO multimedia 3rd Generation 1995 96 Full motion HDTV MPEG-2 MPEG-3 MPEG-4 100 MHz Pentium 2600 MB disk Integration of OO with systems operating system FDDI (500 Mb/s) ATM 4th Generation 1996 On-demand video/film Multimedia videoconferencing Wavelets C ???? ?????????? C Risc processor Integration with corporate KBIS Multi-Megabyte ATM Isochronous Ethernet Technology: Base platform: Authoring: 25 MHz 386 40 MB disk Hypertext Hypermedia Ethernet Token ring Network: FDDI (100 Mb/s) interrelationships are between the four groups shown but some interrelationships exist within groups like the government supporting more than one library and hospital or more than one educational institution at the same time. Whatever happens. and two. or browse through any data/knowledge-base or surf anywhere on the Internet.8 would be one jumble of lines with many being in the north-west corner between the communication service providers and the entertainment industry.Multimedia with telecommunications Table 18. Multimedia applications can be classified in many ways. 1st Generation 1989 91 Media: Text B/w graphics Bitmap images JPEG 2nd Generation 1992 94 Moving.4. One could also do videoconferencing from the desktop and. test beds show no great enthusiasm for such a scenario. These applications are shown in Figure 18. There is the fear that costs of delivery will be so loaded with bells and whistles (features) that are not all needed that it may not be economically viable for many an office and family. through friendly mergers and alliances. There are two crucial characteristics of multimedia: one. This is where the large and very large multibillion dollar companies are trying to stake out their share of the market. Summary and conclusions Multimedia has come a long way since the first generation of applications in 1989 90. Multimedia applications are easier to use and even absorb than many traditional computer 211 .9. have access to knowledge wherever it may be through personalized information channels. digital libraries. and telemetry and other data needed for telemedicine. especially with a trillion bps transmission.4 Evolution of multimedia computing. There are few if any technological roadblocks for the above scenario. This evolution is summarized in Table 18. However. Technology is improving. it will certainly be exciting. or through unfriendly takeovers and buyouts. high sensitivity to delays and lack of synchronization in transmission. all from the comfort of a home theatre. Virtual reality applications will have high sensitivity as well as high volume running into the gigabit zone. These interrelationships if drawn on Figure 18. In between are the multimedia distributed applications especially video/films-on-demand. They have medium to high sensitivity and medium to high volume sometimes going into the gigabyte zone. The end-user also needs confidence in being able to control the content of what comes gushing down the information highway.10. On the other side of the spectrum of computer applications are the non-multimedia applications that have low sensitivity to delays and low to high volume.

Nor can you assign a monetary value to the saving of lives. file transfer Traffic size and volume High VIRTUAL REALITY Low Low Figure 18. e-mail.9 A classification of multimedia Delay Sensitivity High G I G Multimedia distributed applications A B I T ZONE Not multimedia applications e.g. Also important is the application of the digital library and distance learning for they may well save people from remaining illiterate and living on 212 the periphery of society. In this sense the value of multimedia applications will be infinite even if they never become killer applications but become mainstream applications.10 Data sensitivity and traffic size applications but more difficult to produce. The applications of multimedia examined in this chapter are no ‘killer’ applications like say e-mail was in the early 1990s.Telecommunications and networks END−USER to END−USER INTER− ACTIVE ACCESS • Conferencing • Training • Education END−USER to COMPUTER Graphical User Interface END−USER to DOCUMENT Hypermedia access to documents COMPUTER to END−USER Real-time film/video/ shop/bank/ medicine BROAD− CAST Presentation Information kiosk Newsletter Distance Learning OBJECT− ORIENTED MANIPULATION Collaborative processing Data/Oo knowledge− base Document e−mail Tele−medicine Distance learning Figure 18. And what is the value of mass entertainment on demand? This is somewhat questionable because that application has a downside of sometimes getting what you do not want either as violence or pornography or as being . But the application of telemedicine may be a killer application in that its absence may kill people.11. They have special resource requirements that are summarized in Figure 18. The value of spreading knowledge and increasing learning cannot be measured.

28 (5). Businesses will benefit greatly from distributed multimedia applications.11 Resources needed for multimedia undesirable for children. system can substantially improve patient care is clinical neurophysiology.’ MedNet has also developed communication control strategies for a wide variety of data including audio.1: MedNet MedNet is a medical collaborative and consultative system that was in its third phase of operations at the University of Pittsburgh Medical Center. Multimedia MedNet. IEEE Computer. .Multimedia with telecommunications ATM FDDI SONET High thruput rate (compression) High transfer rate (bandwidth) Isosynchronous service TI Workstation (high performance) Servers (special purpose) Peripherals Storage (high capacity) ATM Network Protocols Architecture OSI extended TCP/IP ISDN Hardware Standards H series MEG-I-encoded audio NTSC quality video CD quality MEG-2-encoded video HDTV-quality video IPEG Organizational arrangement Merger Alliance Takeover Figure 18. Document processing through e-mail will improve business communications and teleshopping using multimedia may even increase sales and commerce. Systems Britannica. It is very possible that every worker will have the capability to videoconference from the desk without the elaborate equipment now available only in special rooms equipped for video-conferencing. MedNet provides realtime monitoring and multiparty consultation and collaboration during brain surgery for approximately 1600 cases per year.2: Electronic publishing at Britannica Encyclopedia Britannica and Wide Area Information Services (WAIS) Inc. of the US agreed to jointly develop improved information retrieval for the Internet. Businesses will have to devise new strategies for advertising and selling. During brain surgery. . . . This poses a problem for technology: should technology be responsible for the control of content or should it be the owner of the system (parent in the case of children using the system)? The problem is less a technological that a moral and ethical problem and the response will vary greatly among societies around the world which will make the global use of multimedia all the more difficult to define. p. A major challenge in implementing MedNet was ‘the development of techniques for processing and display of real-time multimodel medical information . a consultative service of diagnostic and monitoring service of diagnostic and monitoring techniques used to assess nervous systems functions . Source: Robert Solomon. the publishing platform that incorporates the new searching technology serves as an infrastructure for Britannica’s electronic products including natural language searching which allows the end-user 213 Case 18. May 1995. MedNet started in 1985 and is on-going on a daily basis at seven hospitals and multiple diagnostic and research laboratories. Case 18. One area where a collaborative . video and neurophysicological data that arise in a medical environment. much less resolve. while consumers will have to adjust to new ways of buying and receiving the products they want as well as the entertainment they desire. 65. The new search-and-retrieval engine optimizes the original WAIS engines producing efficiencies of up to 20% with easier and faster access to the 44 million word base of the Encyclopedia Britannica. monitoring techniques help prevent damage to nervous system structures by continuously measuring activity receded directly from the brain in real time.

Case 18. University of Illinois and at Michigan University. Supporting cooperative medicine: the Bermed project. 65. The cost of such an equipment configuration is currently $50 000 but is expected to drop by over 50% as technology improves and it becomes easier to transmit large blocks of data by wire and fibre-optic cable. Many insurers will not cover telemedicine of this type and many lawyers worry . 214 Case 18. A self-evaluation concludes: ‘The use of multimedia DTC has shown the need for video. Alexandria in California. Case 18. Final acceptance in routine medical practice. Communications of the ACM. 2.eb. however. graphs. film and video sequences. Gary Doolittle. systems do not operate optimally and many systems operate in isolation with little or no integration with other related systems. 1995. with a collection of over 18 million volumes.. Also possible is an elaborate systems of hypertext links that make it easy to navigate among the related entries in the database. Advanced multimedia and communications technology support all areas of medicine from computer tomography and magnetic resonance imaging to intensive care monitoring units and administrative support systems.4: Video-conferencing in telemedicine at Berlin Video-conferencing is part of the Desktop Conferencing (DTC) used extensively in the Bermed Project at the German Heart Institute and the Rudolf Virchow University Hospital in Berlin. charts. p. and conversations with patients and colleagues. damaged by the fire in 1731. The Britannica’s Home-Page on the Internet is: http:/www.’ Source: Lutz Kleinholz (1994).3: The access projects at the British Library The British Library. The Network OPAC have over 6 million bibliographical 59 64.’ Electronic Photo Viewing System has a major photographic collection including Victorian spiritual photographs is accessible by subject and provides a hyperlink to the descriptive text which accompanies each image. Software searches and prints high-quality copies for users in under two minutes. 38(1). audio recordings. ‘Beta sites are now being selected in the US to work with the library on establishing communications links and usage requirements for a future international Network OPAC. talks to the patient and uses twoway television along with stethoscopes.’ Source: Jonathan Purday. including changes in work practices to obtain maximum benefits.Telecommunications and networks to enter a question like ‘Why does the moon loom larger on the horizon?’. However. reference books. gesture support and overall simplicity. X-ray transmission and lab tests that are performed remotely using telecommunications. will depend not only on technological factors but also on the many social and organizational issues. Some of the projects include: The Patent Express Jutebox holds over 34 million patents. a 37 year old oncologist. p. University of California at Berkeley. Diagnostic information comes from a variety of imaging techniques and formats as well as reports. are now discernible and test images have been mounted on the Internet to allow international scholars to see the progress of the project. handwritten notes. The British Library’s initiatives for access projects. high quality audio. The Electronic Beowolf holds the unique manuscript of the 11th-century Anglo-Saxon epic. Source: Information Today. or hidden by 19th-century restoration. published a commitment in its Strategic Objectives for the Year 2000 of ‘providing maximum access to the collection using digital and networking technologies for onsite and remote users’. IEEE Multimedia.5: Telemedicine in Kansas Increasing in popularity in the US is the use of telemedicine as in the case of the Kansas Medical Center where the patient is 300 miles away and the doctor. 1 44 52. Note: For digital library projects at Stanford University. see Ibid. pp. The future success of multimedia DTC in medicine critically depends on interoperability and solutions to security problems. ‘Letters and words erased by the original scribes.

Gemmel. (1992). F. and Marshall. 85 96. professor of bioethics at the University of Pennsylvania. 6(4). IEEE Communications Magazine. and Cukor. International videoconferencing: a user survey. 38(1). and Venkatraman. 38 41. Lippis.E. Snider. 68 79. T. The video communications decade. D.. D. (1994). S. H. D. Journal of End-User Computing. M. (1992). 44 52. A new view of intellectual property and software. Bleecker. B. 28(5). 77 84. Networking multimedia: how much bandwidth is enough? Data Communications. W. 24 33. 104 163.M and Yen. M. (1995). Education wars: the battle over Information Age.. Thuston. Communications of the ACM. and Weiser. Peleg. and Reichman. (1996). value. 17. K. A. 215 . Vin. communications of the ACM. Multimedia: the challenge behind the vision. K. Multimedia. Rangan. 12 23. Levy.D.Multimedia with telecommunications about malpractice and privacy.. Multimedia networking. Datamation. 74 80.. J.J. Domet III. Data Communications. Bibliography Akselsen. P. An opensystems approach to video on demand. 47 58. Telecommunications. 28(5). Coopock. 21(12). The Futurist. 1(3)..D. 63 64. 14 23. Multimedia information systems. Verbiest. Data Communications. L. Intel MMX for multimedia PCs. D. D. Multimedia networking: an applications portfolio. K. 22(2). 10(19). January. . 28(5). (1996). J. 85 94. (1993). W. R. IEEE Computer.L. J.. as medicine moves into cyberspace. May/June. Multimedia pedagogues.H. (1994). Rajkumar.R. P. Video teleconferencing: the state of the art. (1995). (1995). Telemedicine and ISDN. Little. and Rowe. 103 111. Kandlur. N. 32(5). Multimedia storage servers: a tutorial. Ninke. (1994). Towards vision 2001: voice and audio processing considerations. IEEE Communication Magazine. Rodriguez. Resource management in networked multimedia s1562 ystems.. and Folkow.E. U. 1(1). telemedicine is already here’. Multimedia networking. W. Al and A10. McQuillan. (1995). (1992).H. Information Systems Management.. Communications of the ACM. AT&T Journal. 82 88. D. but that overcoming the economic barriers ‘may take 20 more years . (1994). W. The important resistance comes from local physicians who say that they are afraid of out-of-state quacks that are not governed by state licences but actually fear the economic threat to their practice of clients that can now use telemedicine to go across the state borders (and maybe across national borders) to seek medical treatment or at least get a second opinion through a televideo telemedical appointment with a world famous medical specialist. Furht. W. Doctors can Diagnose Illnesses Long Distance to the Dismay of Some. and Verhille. (1993).. 25 39. (1995). Tobagi. Beyond talking heads: videoconferencing makes money. Reisman.. Pitt.P. Computer. R.J. Design issues for interactive television. L. (1994). P. Kitson. C. (1993). 46 58. Davis. Rabina. Deloddere. 28(5). 4 13. 32(5). 21 30. 26(4).V. 33 34. Richardson. Kapor. (1994). D. telebanking and securities trading. Wall Street Journal. Coggins. Futurist. 30(3). J. 26 30. The emerging meta-market. (1995). (1993). Aldermeshain. argues that ‘technologically.A and Wall.H. Prospects for interactive video-on-demand.M. (1994). and Venkatash. Telecommunications. 24(9).C. 38(1). 40(1). 31(1).K. Computer. B. Digital libraries. 11(4). pp. S. Interactive video on demand. Source: Bill Richards. T. Telemedicine is thus becoming another form of electronic commerce much like teleshopping. Strauss. Woolfe.. 17 19. C. Videoconferencing: the bigger (and better) picture. D. 40 49 Grosky. T. 1996. and Pile. Eidsvik. S. S. J. IEEE Computer. F. 2 6. (1994). H.J. Chang. Wiederhold. Multimedia 1(1). D. and Hall. G. 61 64. 74(2). IEEE Multimedia. B. (1994). and productivity.. 14(7). 72(1). P. (1995). 60 69. Shot by shot.M. 52 63. (1997). Jan. A. AT&T Technical Journal. Furht.. Kalia. Tolly..A. PC Magazine. Multimedia Computing: an overview. A. S. (1995). (1995). Nahrestedt. Skellern. 24 30. Arthur Caplan. 39(3).M. Going digital: a look at assumptions underlying digital libraries. IEEE Communications Magazine. . Y.J. 22(3). H. Ozer.. by eliminating the middlemen and bringing new competitors into previously insular markets. Data Communications. Samuelson. D. 39 45. (1995). 32(13). Wilkie. Thapar. Communications of the ACM. Multimedia systems: an overview. we need a new system of checks and balances’. Data Communications.

There are other names for this type of worker at home: electronic homework. The emphasis is variously on work being independent of place and time. Telecommuting Knowledge workers work mostly in offices but some work at home using telecommunications to keep in contact with their corporation instead of being in the office physically. This third condition is largely satisfied with many in the workforce who are computer literate and comfortable with the PC which is end-user friendly. But this perception does not hold for the teleworker. Stewart The general idea was that the industrial revolution had taken people out of their homes. or international connecting the whole world. knowledge workers have increased in number and importance. Tom Forester Introduction Knowledge workers handle knowledge. but knowledge in the sense of the basic units of data and information. The fourth condition for telecommuting is that there are professions where telecommuting is feasible. customers. suppliers and other business players which is supplied by telecommunications and networks. or telecommuters. With the confluence of computers and telecommunications. where all the teleworkers are men. International connectivity is currently achieved through the Internet and e-mail. at least not at Rank Xerox. distance workplace. These workers are 216 called teleworkers. and especially telecommuters. a connection of this device with the corporate office. All knowledge workers. E-MAIL AND INFORMATION SERVICES Networks connect people to people and people to data. and three. With the coming of computers. fast and robust computing device that is satisfied by the PC or workstation. that is. need interconnectivity. a cheap. largely on economic grounds: corporations do not have to pay for the high overhead of an office building nor for parking and other . their computers must be connected to other computer systems. a worker at home who is at ease and comfortable with a computer and telecommunication. There is a common perception that many teleworkers are mothers who stay at home and join the workforce to work around feeding their children and doing their household chores. telecommuters are increasing in importance and numbers. This interconnectivity can be local within an organization. in this chapter we discuss e-mail. two. virtual work. One subset of knowledge workers includes those that work at home and they are referred to as telecommuters. and now the telecommuting revolution will allow them to return. But why is teleworking a relatively new phenomena? Because it has at least four prerequisites that have to be satisfied: one. The Internet we examine in the next chapter. or regional within a community. flexiplace and electronic cottage. viable and desired by corporate management. In this chapter we examine what knowledge workers and telecommuters do and how they do it. though perhaps not as end-user friendly as some would like it to be. Rank Xerox in the 1980s called it networking. This condition too is often satisfied.19 TELECOMMUTERS. Thomas A. not necessarily in the sense of Artificial Intelligence where knowledge includes heuristics and inferences.

The teleworker argues that in return for not being paid like full-time employees that work in the office. whilst dropping them when demand drops. Telecommuting reduces the costs of carrying employees not needed all the time and the cost of fringe benefits that must be given to the full-time employees. auditors. the teleworker does not have to dress up for work and drive to work. its servers and the corporate database. memos. Outside IT. phone. regularly scheduled meetings. It was then recognized that roughly 5% of the US oil was for commuting and that a savings of 190 000 barrels of oil a day could be gained if 20% of the commuters stayed at home. telecommuting is viable for sales persons. Once connected to a corporate LAN. telecommuting offers corporate management flexibility in the hiring of workers when needed. Telecommuting is the substitution of telecommunications and computers for transportation to the office. Such an interconnection through the Internet is shown in Figure 19. reservation clerks. Other organizations design work around assignments so that employees who operate out of their homes spend one or two days each week at the main office. training should include ways to remain in contact through electronic mail. Sometimes a company will establish a small satellite office with an electronic link to the main office. Technical or documentation writers and IT training specialists who must develop educational materials are also good candidates for telecommuting to do part of their jobs.2. The percentage of time an employee telecommutes varies from one organization to another. When teleworkers are not union members. Informal meetings take place in low rent neighbourhood centres or at the homes of employees. By that time reports on the success of experimental alternative work-site programs were in circulation and the benefits of telecommuting were better understood. They can work when and where they please. The people selected for telecommuting must work well on their own. For complex tasks it may be necessary to have a workstation equipped with peripherals. It was first talked about during the Arab oil embargo in the mid-1970s. but not till the 1980s did the idea come of age. Persons selected for telecommuting should be trained to cope with off-site employment while managers should be trained to supervise subordinates from a distance. routine 217 . another factor contributing to the economic feasibility of working at home. fax. On the contrary. Some companies have no main office at all. all without any cost or firing or any hassle with the unions. There is also greater flexibility in hiring because telecommuting adds to the labour supply. though when not in the office they miss the socializing and time off they sometimes get in the office. Implementation of teleworking The equipment required by a telecommuter could be quite modest: a terminal with a modem. architects. Their assignments must be jobs that can be done away from the office. programmers. the unions are unhappy with telecommuting and are often actively opposed to it. A few corporations did experiment with home-workers at the time. Thus. and even some systems engineers.Telecommuters. This part-time arrangement has tax advantages in some countries and is an opportunity for people who want to stay at home and still be part of the productive workforce adding to the GNP (Gross National Product) of their country. If the corporate LAN is connected to the Internet (which is a net of other nets) then the telecommuter can access any customer or vendor (or friend and relative) anywhere in the world and at any time desired.1. This type of alternative work style is classified as telecommuting since the time and cost of commuting to an office is avoided. Unions argue that teleworkers are exploited with low wages and no fringe benefits compared to what is paid to fulltime employees for the same work. the personal computer was on the market and at an affordable price. Often though. accountants. the peripherals run by the corporate computer system can be used. IT personnel that include data preparation clerks. All work is done at dispersed workstations joined by a network. e-mail and information services facilities for its telecommuting employees. In addition. are good prospects. some secretarial positions and other white collar professions. Remote work does not mean all contact with co-workers is lost. Careful planning is required for telecommuting to be successful. Such a system is depicted in Figure 19. the unions are unhappy in losing membership and the clout and power associated with large memberships. the telecommuter can not only have access to expensive peripherals but also to the main computer system. Thus. Telecommuter applications can be stored in the database and called upon when needed as part-time employees for the duration needed.

clothing. meals. but the rest of the . But many evaluations are often qualitative and subjectively judged by observing the workers attitude and behaviour at work.) and do not realize the problems associated with working alone. There is some evidence that persons that opt for telecommuting look only at the short-term benefits (less commuting and reduced expenditures for petrol. to an extent. For one. how do you evaluate workers with whom you have no (or little) eye contact and do not observe while at work? Work performed at home by telecommuting can be evaluated quantitatively and perhaps.1 Possible access to equipment by teleworker Our National WAN Our National LAN Teleworker's corporate LAN Internet MANs abroad WANs abroad LANs abroad Figure 19. there are new 218 problems for management.Telecommunications and networks Image Processor Database Server Data base Computer Corporate System To other LANs Optical Scanner Application Server Corporate LAN Videoconferencing facility Printer Client for Teleworker PC & Workstation clients Copier Figure 19. etc. Besides the economic. qualitatively also. A training period can help employers evaluate whether the right candidate for telecommuting has been chosen and help employees gain a realistic view of telecommuting as a work mode away from the social contacts of a city and an office. This is not possible in telecommuting and new ways of evaluation have to be devised appropriate for the tasks and personnel involved. social and technical considerations of telecommuting. and so forth. This may involve some face-to-face meetings.2 The Internet as a net of other nets reporting.

Many employees today resist relocation. instead gained by telecommuting. the employer avoids the cost of the relocation or the cost of hiring a replacement. For example. C. discipline and planning skills to direct their own activities. In addition. However. Companies that have seasonal peak volumes or certain hours of the day with Corporation Figure 19. Persons with medical disabilities who have difficulty leaving home. Gains in productivity of 20 to 40% are not uncommon among professionals. J. Organizations trying to hire professionals when their demand exceeds supply.3 Advantages of telecommuting 219 . uses telecommuting for some of its catalogue-order telephone centres. Telecommuting can also be an alternative to corporate transfers. employees at home can take orders and answer customer queries.Telecommuters. In service-oriented organizations. it should be recognized that many persons work best in a structured environment or when they interact socially with others. e-mail and information services evaluation must be based on measuring and monitoring productivity at home. Cost savings Productivity increase Recruitment and retention of staff Morale Productivity Opportunity for disadvantaged/disabled Teleworker for the geographically isolated Flexitime Working environment more 'friendly' Combine work with leisure Quality improvement Creativity enhancement Less pollution Rural development Urban development Society Energy consumption Work-force Unemployment Less transportation Fewer offices Fewer car parks The benefits and limitations of telecommuting One of the primary benefits of telecommuting is increased productivity. will find that the option of telecommuting gives them a recruiting edge. a large retailer in the US. Careful screening of candidates is necessary to ensure that productivity is not lost. a valued employee who might resign rather than move can be retained when telecommuting is an alternative to transfer. can now enter the work-force through telecommuting. Monday to Friday work week. When working at home employees can choose their hours of work. Telecommuting is also attractive to persons with family responsibilities (a young child or aged parent) who may be unable to work at certain hours of the day or cannot take the time to commute. Not every one has the work habits. telecommuting can help attract applicants. A common reason is that a spouse is employed and unwilling (or unable) to uproot. telecommunications is a good way to extend hours of operations. or who just do not like to commute. Many persons like the idea of flexible scheduling instead of the standard 9 5. Whatever the objection. Penney. Workers claim that there are fewer distractions from the central office. heavy demand can handle the load by telecommuting without tying up office space. When organizations are recruiting new employees.

When telecommuting? When the advantages can overcome the disadvantages and limitations of telecommuting. 28 9) include: people who wish to work at home in order to avoid the commute to work. there are many people who would want to telecommute. disabled people. Expands the workforce to: housewives. Organizations can expand rapidly without worrying where to put people or retrench without being saddled with expenditures for unused space and furnishings. 220 . Despite these reservations. women who wish to stay with their children at home and yet want or need to work. people who feel that they are exploited and underpaid at work. telecommuting means that space is not a constraint when changes in the workload are proposed. networking (local Finally.1 and Figure 19. firms and organizations that find telecommuting a way to downsize their organization and find telecommuting of some of their employees an economical thing to do. women who cannot find child care arrangements.Telecommunications and networks Table 19. There were some 50 videotex’s operating in 1982 in 16 countries and all of them operated below expectations and often below the breakeven point of profitability. workers who are ‘loners’ by temperament and rather work at home than in an office. Psychological and social needs of teleworkers may not be met. Provides access to ‘piece-work’ and part-time work. people in remote regions and even farms. people who use computers at work and might as well use them at home. Offers teleworkers the flexibility of working at their choice of time and place. work conditions. Most predictions in some countries went wrong partly because the predictions were made by vendors and vendor-biased commentators and partly because the predictions were based on poor assumptions of the behaviour of parties at home and their ready acceptance of computer technology. direct and indirect. Keyfax in Chicago. payments. AI technology (including expert systems).3. Inability to separate work and pleasure. Possible ‘burn-out’ of working at home. Raises issues with unions over: membership. Limitations and disadvantages Raises problems of: providing proper access to information needed. there are some predictions that can perhaps be safely made. individualists who find the large corporation so antipathetic an environment that they wish to set up their own office at home. Viewtron in Florida. integrity and security of data/information. people who want an extra job.1 Advantages and disadvantages of teleworking Advantages Offers flexibility to corporations in the hiring of transient personnel for peak and unexpected loads. The home-office of the teleworker will soon be the confluence of computer technology. Furthermore. evaluating teleworker. Conflicts and ‘over-crowding’ at home. The success of Minitel was due partly to high end-user acceptance but also to a combination of political intrigue and the deep pockets of government support. These people (see Ursula Huws (1991: pp. There was one exception: Minitel in France. telecommuting is advantageous because it reduces office space and furniture requirements. Future of teleworkers Before making predictions about teleworkers. The advantages of telecommuting are summarized in Table 19. one must remember that predictions of the use of computers and telecommunications in the home have been made in the past and have been notoriously wrong. Gateway in California and Bildschirmtext in Germany. Witness the less than expected use of home shopping and home banking through videotex both in the US and in Europe: Prestel in the UK. people with disabilities but who can work at home. Provides a possible ‘creative’ environment of the home. With telecommuting such overhead expenditures are minimized.

wire-less technology. They will have the ability to process not just data but knowledge. For local communications a local information service provider may 221 . An important resource for any knowledge worker and teleworker is to have the ability to receive and send messages quickly and easily. the post office and face-to-face discussions. We then examine the resources needed for e-mail. and not just on-line but in real time too. new standards for work and performance must be instituted and monitored by corporate management. This requires a corporate plan for the migration of work that was done in traditional ways to the newer ways using the latest technology both efficiently and effectively. signalling over fibre optics. e-mail and information services Supercomputers Parallel processors Specialized computers Mainframe Minis To (other) LANs and the internet L A Specialized workstation N Your PC/workstation as typical for a teleworker Laptop portables/PADs PC connected to Internet Wireless systems Figure 19. One such scenario is shown in Figure 19. Bureaucracies will be reduced and hierarchies will flatten out. PADs and the Internet computer). There will be the capability of faxing documents and using palmtop computers when needed. restructuring and re-engineering. The telecommuter will have access to many different peripherals through a network that are not available at home.Telecommuters. This is achieved by e-mail provided both parties have an e-mail connection. The changing technology and changing work environment of the ‘electronic briefcase.4. The worker is now liberalized and dispersed without set locations or schedules. and. the roles and tasks may require designing. supervisory roles and leadership roles redefined. In many cases bureaucracies and hierarchies will be replaced by adhocracies because there is little time for critical information to go up and down the rigid chains of hierarchical command and percolate slowly through the many layers of bureaucracy. devices will be ‘smart’ because of their use of AI programs. Access by telecommuting may even extend to certain professions such as the salesperson who works out of a van or car. employees must be educated and trained for the new work environment. All these. the telegraph. or at least most of these. E-mail We start with a definition of e-mail and compare it with other means of communication like the telephone. will require organizational responses. There will be continuous contact between management and customer and back to managers making strategic decisions. for example. image processing and the fast processing of massive volumes of data and knowledge. Some workers may well be displaced from their long-held jobs if not unemployed. not just discrete variables but ‘fuzzy’ variables.’ the electronic office on wheels or at home.4 Interconnectivity for a teleworker and global). often through the intermediary of the teleworker. sometimes not so. These are no great predictive leaps into the future but merely a projection of what is technologically possible and what is being currently done (as. most important perhaps. The electronic office on wheels will enable the salesperson to give video and animated graphic demonstrations to the clients at the client’s site and respond to queries that will bring instant responses from the decision-maker in the corporate headquarters. For many workers.

If a printer is available. But in principle. e-mail is often not charged by the length of the message. Of course you must follow the protocol (procedures) required for the transmission of the message and you must provide the address of the destination according to a prescribed format. press a few M. Some addresses have more subdivisions separated by dots and international addresses may even have a country alphabetic code at the very end. you may use the DNS (Domain Name Service) format which translates the address provided by the user to the numeric address used by the machine. This is somewhat like the telegraph communication which is also swift once you have composed your message. we go the opposite route to e-mail: we start with the detail name. you will get a message from the e-mail postmaster explaining the reason for non-delivery like an unknown computer address. unlike the telegraph. E-mail is electronic communication. There is no dead letter box for e-mail as there is with the post office. and then forwarded through more sorting desks (like in post offices) before they are delivered at the address where they can be collected when convenient. E-mail increases secretarial performance by a factor of two to three. that is. The great advantage of e-mail compared to face-to-face meetings and telephone calls is that you have time to think of an answer. In the postal service of many countries. For example. In addition. the two people involved. Retrieval and the file management capabilities are particular strengths. the sender and the receiver. the cost is not as prohibitive as that of the telegraph. There are at least three other classes but we shall not be concerned with them except to recognize their existence in case they turn up sometime. street and then the number of the house. The process is also costly. K. The transmission in e-mail is swift and very much faster especially for international communication. You do not stand the risk of being misunderstood as in a fact-to-face relationship resulting in unintended emotional responses.’ separating different fields (parts) of the address. A document . Both use the store-and-forward technology. Another commonality of e-mail with the post office is that both are asynchronous processes. and respond when you are ready. You can break off at any time by simply not responding. do not have to be available at the same time. The delivery point may be the home PC (or 222 workstation) or the office. you start with the area code. You are never sure of delivery in surface mail (unless you get a reply) especially with foreign mail for you do not know if the postman has taken a liking to your stamp. both locally and internationally. The computer system uses a numeric address but that may be difficult for the average user to remember and so the address you provide is often alphabetic with some special symbols and the systems protocols does the translation. provide the ‘address’ you wish to send the message to. Mailgram and fax are alternate methods. To communicate across the world. name of the city.Telecommunications and networks be sufficient. After the name. It must also be recognized that e-mail is just one of the many modes of electronic message transfer. as is the case with a telephone call or a face-to-face meeting. We shall briefly discuss the Internet as preparation for our discussion of cyberspace in the next chapter. and. The e-mail address is somewhat like the addressing of an envelope in Germany. You can take your time. are delivered to a local sorting centre where they are sorted for the address of the destination. like your letters. Hussain at (@) AOL (the information service provider) and com is the short for ‘commercial service’). This address you provide starts with the ‘screen’ name of the receiver. there is the Internet. One advantage of e-mail is that when sent you know it is delivered and so if for some reason it cannot be delivered. taking between 1 to 6 minutes to transmit a doc2 ument copy comparable in quality to a Xerox copier. This is just one class of addressing.e. Fax currently is slow in transmission. Thus the author’s e-mail address is: KMHussain@AOL. You are not on the spot (as with a telephone call) and under duress to express your opinion instantly and seem intelligent and profound. but. if it is so charged. and the message is sent to the desired ‘address’. then the network ID and working towards the finest locational information with the ‘@’ sign and dots ‘. You enter the message you wish to send on a computer. Your email messages. And it is much cheaper too. i. both incoming and outgoing messages can be printed in addition to the soft message on the screen. the system reduces the need of office storage since cabinets with files correspondence can now be replaced with computer memory. street and then city. e-mail is much like any post office.

Whatever the approach to connection. MIME is 223 Resources for e-mail There are three important resources needed for e-mail: a computer. miss the human interplay of visual messages or a phone call or a visit to an associate. and 60 times that of a typical office memo. STMP is Simple Mail Transfer Protocol. how you say it and when you say it. it is TCP/IP (Transport Control Protocol/Internet Protocol). Get an Internet connection through an Internet gateway which then enables you to do e-mail not just nationally and regionally but also internationally. you cannot see emotions in a blush of embarrassment or of joy. there are several alternatives: TCP/IP. independence. the MIME. In the case of the Internet. That is 1000 times the number of bits required for a typical telegram. and for the private e-mail system. PPP. It is also used by the Internet and is simple in the sense that it handles simple messages of text and numeric data. 2. But it is such private systems that only the large organization can afford or indeed need. 3. This private systems approach is more expensive but it allows customization. and you cannot sense the atmosphere that may be tense or frivolous. then it can be added on. But you have total control over what you can say. you cannot express joy or sadness through a change of tone or inflection. You have no physical clues of how others are thinking or reacting to what you have just expressed. as well as receiving and responding to messages when convenient. Essentially. You cannot hear the sounds of anger or pleasure. one wants to transmit pictures. Build your own e-mail backbone with a dialup access and also access to the Internet. You cannot interact. SLIP. . a modem allows us to convert the digital data of a computer into analogue data for the telephone and vice versa. A modem stands for modulator and demodulator and is used for communication by a computer (digital) over a telephone line (analogue). There are at least three possibilities: 1. This is the demodulator part of the modem. sound and video. These are LAN based systems and the most cost-effective for most users. It is a device that transforms a computer’s electrical pulses into audible tones for transmission over the phone line to another computer. If not. A modem often comes built into a computer. there is PPP (Point-to-Point Protocol) or SLIP (Service Line Internet Protocol) connection. Get connected to an Information Service or Business Communications Service provider that offers e-mail as just one of its services. If. You are alone. then one needs an additional protocol. But for these and the other advantages of e-mail. retrieval and cross-indexing capabilities. better administrative and management control. you must pay a price. one must use a protocol for messaging. Email often eliminates the need of voice and personal contact in business. Multiservice Internet Mail Extension. being shielded from constant interruptions by jangling phone calls. In addition you have all the facilities of the Internet. especially the gregarious ones. We discuss such a connection and the Internet in general in Chapter 20. But e-mail demands a change in style of management. This is the modulator part. a modem and a connection to a e-mail provider (who provides you with the software necessary for the connection). Besides a computer and a modem. e-mail and information services page in facsimile form requires 200 000 bits. In spite of its obvious advantages. Let us consider the modem though it has been mentioned earlier. and even better performance and scaling than the shared public systems. however.Telecommuters. Business people. This will be considered a private network as distinct from the other two alternatives which are public backbones. what is also needed is a connection and the software to be able to do e-mail. We shall examine such providers and their services later in this chapter. for the information service provider. Do not let the computer buff confuse you with bandwidths and megahertz speeds for they are significant only if you have large volumes of data or are transmitting voice. or STMP. corporate managers tend to be ambivalent in their attitudes towards electronic mail. You need computing resources. images and sound which you will most likely not be doing if you need to know the basics about a modem. E-mail has its drawbacks especially when compared to face-to-face meetings. The modem also receives incoming tones and transforms them into electrical signals that can be processed by a computer. They like the speed.

this might result not just in better and faster communication but perhaps also in the flattening of the In Europe. For example. There are some programs around that will filter your mail. SWITCHING METHODS. document sharing. in the US. This hierarchy is illustrated in Figure 19. The filtering and screening of e-mail can EDI E-MAIL INFORMATION RETRIEVAL MANAGED DATA X.656@compuserve. or it could filter by subject matter according to key words that you provide. They lie on top of what can be looked at as a cake with e-mail and EDI being the top layers. however. the X.5.400 (and X. as well as better directory synchronization. They must thus seek the package that serves their needs best and is most cost-effective. E-mail may yet become an important enabling technology for enterprise computing. there is greater respect for the ISO standards. faster offering of products to the market. and increased productivity. It could check the address of the origin against a list that you approve. Having just mentioned EDI. changes in work structures. directory services and e-mail APIs (Application Programming Interfaces). as in the DSS and EIS. For corporate management. The X.25. there is the X. it may be the right time to establish the relationship of e-mail to EDI. The X. There is the need in e-mail standards for file formats.500 that is more widely used and is the directory standard for the X. group work and ultimately decision-making. the bottom most layer.435 for EDI on X.400 is the alternative to STMP as a mail backbone and it is also the standard for EDI in both Europe and Japan. In Europe. at least the bottom most layer. Creating and maintaining comprehensive lists poses great technical and organizational problems when the scope of the addressing is not just national and regional but also international. There is already an acknowledgment that a larger width of the fields for addressing will be necessary for any future global messaging system before we can link heterogeneous networks and synchronize distributed directories of networks and e-mail users.400). One annoying problem faced by e-mail users is the volume of junk mail that comes through. which may well consume around two hours just in reading time. the improvement of control and tracking. transports. it is easy to get 200 300 messages a day. If you are on various lists for mail. with managed data on top of the lowest layer and information retrieval on top of that. What must be deduced from the cake structure is that you cannot always just simply choose one layer and not the bottom layers. LEASED LINES Figure 19.400 was prepared by the ITU (International Telecommunications Union) and endorsed by the ISO. And so the consumer is sold a package with layers that they may not want or need. you may have an alpha address (with an ‘at’ and ‘dot’ symbol) and yet a relative or best friend may have an alphanumeric and symbolic e-mail address like 76361.Telecommunications and networks the protocol for multimedia e-mail and can be used also with STMP. The protocols mentioned above are most common approaches for the US where there is a spectrum of vendors and corporations that are large enough to go their own way and to embrace a message handling system that includes e-mail as just one of the many message transport systems. switches and leased lines. This will greatly help in building the messaging infrastructure which would facilitate not just email (and fax) routing but also database access. An irritation for the consumer and perhaps the greatest obstacle for an easier and cheaper system is that there is no common global directory or universally accepted standard for e-mail addresses.5 E-mail in the hierarchy of layers 224 .

as the parties involved double in number. features of end-user friendliness. Some are good at e-mail and others are not. cost. news local. One study showed that. membership services. travel. sport. GUI. e-mail and information services control the stream of e-mail but that will greatly depend on the updating of the lists used for the filtering. leisure. Different providers have interfaces with varying degrees of end-user friendliness. As the total number of computer users increases and the percentage of users increases. ease of use. the messages increase fourfold. installation. But e-mail is not the only service that the information service provider provides. hobbies. Criteria for such a comparison and evaluation could be the following: ž ž ž ž ž ž ž ž ž ž ž ž ž ž ž ž ž content for areas of interest. business careers. In scope. E-mail services are often offered through a GUI. from access to databases to collaborative applications. an Internet provider or a LAN. Summary and conclusions E-mail is fast. ease of installation. e-mail fee. interface. Its primary function is to provide information on subjects like ž ž ž ž ž investing and finance. that is menu driven. graphical user interface. Despite problems with e-mail it is growing rapidly in volume and scope. local/long distance affecting costs. Internet connectivity. entertainment. shopping. speed of access affecting waiting-time in queue.Telecommuters. weather. for example a scissors icon to cut a selected passage). it is extending from local to global communications. reliable and cheap if you have a connection to some service provider. and end-user friendly. Information service providers The tricky question now arising is to find an email provider among the many information service providers. type of ‘community’ in chat and discussion groups and forums. And this is with only 42% of the 47 million computer users in 1993. e-mail. 1995: p. e-mail in the US has risen over 60% from 1992 1993 and from 5. from sending mail to customers and suppliers to sending mail to friends and family.9 million in 1992 to over 38 million in 1995. billing services. ability of parents to control content of what children and teenagers may watch. e-mail will also increase. E-mail is becoming increasingly central to the way we work in the office and at home as a teleworker. if any access. It is also central to a corporate messaging system. Each should be evaluated for appropriateness to one’s environment. national and international. Such a provider is the equivalent of the post office that does the storingsorting-and-forwarding of messages and provides you with an e-mail address and the software to receive and send messages. there is a good chance that e-mail will continue to grow and prosper. features such as address book and its use in composing mail. E-mail capability can be acquired through an information service provider. In numbers. games. Early systems had a command driven language but nowadays you can get a GUI (Graphical User Interface) which is menu driven and has icons (graphical symbols that perform functions. E-mail is only one of the services offered. ž ž ž ž ž ž ž health. 108). graphical user interface. often in colour. fixed or variable. from a formal correspondence vehicle to a e-mail transport in the enterprise-wide messaging infrastructure. Thus the intracompany and intercompany messaging increases rapidly (Burns. 225 . from business correspondence to electronic form routing and process automation. Will it increase faster than our capacity to fully use it? Or will there be a backlash from the inefficiency and unfriendliness of the system? Time only will tell but the chances are that with the awareness of the problems and attempts to overcome them. from people to people to process to process (virtual users).

car parks. Just as the industrial revolution has brought the worker out of the home.2. The greatest impact will be on society if it transforms working in the office to telecommuting and working at home. the new information era of telecommunications and computers have brought the worker back to the home. One classification is shown in Figure 19. offices. a long time commentator on telecommuting: . 226 Table 19. Despite its limited scope. motorways and highways. and pollution. there will be a great demand for the knowledge worker. There are many types of teleworkers.6 Classification of telecommuting As information processing and knowledge processing increases.Telecommunications and networks at home By location at site (mobile) centralized By organization decentralized satellite centre neighbourhood centre corporate LAN on-line Internet By equipment workstation other printer PC off-line Figure 19. Not obvious are the very intangible benefits to society in the form of better and cheaper urban development with fewer high rise office buildings. fewer car parks. and service sectors of the market economy. and fewer roads. transportation. But this is true only of certain professions and certainly not true of the agricultural. like mothers and the disabled Requires a new discipline for working at home ON THE CORPORATION Enables transient and piece-time employment without the burden of fringe-benefits and working with unions Reduces overhead for office Requires new approaches to scheduling work.2 Impact of telecommuting ON THE INDIVIDUAL Allows employment for part-time work Allows employment for those who do not want to or cannot leave home easily.6. We conclude with some thoughts by Ursula Huws. The impact of teleworking is summarized in Table 19. evaluation and control of remote work ON SOCIETY Reduces pollution from traffic by office workers Reduces urban development and transportation infrastructure Increases potential labor force and potential GNP Makes unions unhappy by decreasing their potential membership. but could also affect town planning for banks. This will not only change our life-style. telecommuting has raised many new issues and conflicts and will need new management strategies ways of resolution. Many of these knowledge workers will be teleworkers. manufacturing.

’ ‘I’ve gained forty pounds. They meet face-to-face every six weeks or so. Some fifteen years later.Telecommuters. 21) Thomas Hubbs Thomas Hubbs has been called a knowledge worker. My terminal is too close to the refrigerator. All parties agreed. There is no incentive to work hard. Case 19. So the firm made another offer. 1993: p. had 100 managers working on trial as teleworkers. (Huws. 1994: pp. p. a telephone company in the US. Dr Bacon now has a two-room office and lots of ‘bells and whistles’ on his office computer system. no sense of personal accomplishment. Verifone’s e-mail system runs on a VAX minicomputer. no feeling that I am part of a team that is accomplishing something. a fax machine. but Mrs Bacon still had to make a sacrifice: she had to give up the bedroom of the fourth daughter for an office.’ (Huws. will depend on the decisions taken by a range of social actors large employers. new ideas of conflict will arise and. They will interact with each other to produce new and unexpected patterns. creative individualists. two telephones. However. 17. ‘everyone from the chairman to the most junior clerk is issued a laptop computer and is expected to learn how to use it. paper memos are banned. clothes. 1994. the Bacons are still happily married and Tom is still a teleworker.’ (Business Week. He rarely uses his cellular phone to ship faxes or e-mail because ‘right now the costs are so high it doesn’t make sense. Ursula Huws Ursula Huws is a freelance writer and writes on telecommuting including the book Telework: Towards the Elusive Office.’ 227 . Case 19. . March April. These decisions will not be unidirectional. entrepreneurs. Hubb’s nominal headquarters is in California and his CEO lives over a thousand miles away in Santa Fe. In 1994. my bed. six book cases. The two parties agreed on all contractual matters but there was one hitch: Mrs Bacon was not in agreement. . . five filing cabinets. ‘three computers. nor will they be permanent. secretaries a rarity at the company’s offices are prohibited from handling an executive’s e-mail. shoes and household linen and. except in an emergency’. 19. and the specific forms which that reality takes. and planners. 95 6) Tom Bacon Dr Tom Bacon. Bell Atlantic. women with dependants. the number grew to 16 000 and Bell Atlantic is working with its unions to increase that number to over 50 000. was offered a job with a software house in the East of the US. an instructor in Operating Systems in California. 13) Hubbs spends 80% of his time on the road often getting behind the walls to link his HewlettPackard laptop computer to an outside line. At Verifone. He is the Vice-President of Verifone which dominates the market for credit-card verification systems used by merchants. a telecommuter and a virtual worker. Internally.2: Comments from teleworkers ‘My professional and home life have become so intertwined that I can never get away from work. Dr Bacon had to come three times a year to the main office for face-to-face meetings and the rest of the time he could stay at home and use teleprocessing for work. She works out of a room at home that has. 29). four chairs.1: Examples of teleworkers Telecommuting at Bell Atlantic In 1991. above it. no social contact. (The Futurist. new social forms will be negotiated. according to her. Oct. in the resolution of these conflicts. 1991: pp. two desks.’ ‘I feel isolated at home. two printers. e-mail and information services The extent to which the electronic cottage becomes a reality. nor did she want to pull her children out of schools in California. She did not want to go to the winters of the East.

2 million Americans will work at home at least part of the time (although 12. And so American Express started its Project Hearth. the kids. Send as many messages as makes sense and this will make your colleagues think that you are working in the office. ‘Remind your boss that given fax machines. January 1995. 5. .’ ‘There have been no raises. studies showed that a typical agent was able to handle more calls at home than in the office resulting in a 46% increase in revenue from booking. Get remote control software that allows your modem to transfer files between computers using phone lines. Source: Fortune. emphasize how the company will profit from your time in the home office. ‘The number is expected to grow by a whopping 15% a year. Source: Preston Gralla. 7. . pp. Case 19. 56 million people will be working at home. the neighbours. ensuring tight control. 8. Of the people surveyed. One such study concluded that no single core of tools and platforms for teleworkers is realistic since there are . AT&T declared a Teleworker Day for all its professional workers.’ Source: Preston Gralla.6: Teleworking in Europe As part of the RACE programme. ComputerLife. modems and communication technology. During working hours. Over 25 000 people worked at home. then there is great potential for distraction and destruction. so by 1997. you plug your notebook into the docking station. When you come in to work 228 1. . 107. The Link Research firm estimates that 43. For hardware.4: Advice from teleworkers The first and maybe the hardest is to ‘try and convince your boss to let you try it . no promotions. pp. Case 19. Get a notebook with a docking station. Jan. 6. and gives you expansion capability. Case 19.’ 4.’ Get a network administrator.’ ‘I’m getting claustrophobia (we simply lack the space) and my wife doesn’t want me home during work hours. even my spouse who knows better I am just too available. half of its 123 000 US managers will try working from home by the end of this century.Telecommunications and networks ‘Everyone interrupts me the dog.5: Teleworking at AT&T In 1994.3: Telecommuting at American Express In 1991. 70% said that they were more productive working at home as opposed to working in the office. 106 10. you can be as hooked in at home as you are in the office. since I’ve had a home office. Case 19. management at American Express Travel Services. Cats have been known to pounce on the keyboards and dogs have chewed up many a floppy disk. 1995. The docking station stays on your desk at the office. the postman. get a system for ‘all home and office work . the EU has been promoting a series of studies to stimulate transborder teleworking.’ 2. p. In 1993. provides power. The big question was the effect in productivity. Supervisors retain ability to monitor agent’s calls. . Work @ Home.7 million of that are selfemployed). Autumn 1993. According to plans of AT&T. which eventually involved 1000 travel counsellors (10% of all employees) in order-entry at 15 locations. 9. ComputerLife. I’m afraid that I’m jeopardizing career advancement by being a teleworker. check and respond to your e-mail at least once every two hours. ‘Work @ Home’. If you are an animal lover. Don’t stress how it will benefit you. and it hooks up to the network. Its many woman travel counsellors wanted to work at home instead of having to brave the driving (sometimes three hours) to work and back each day. faced a problem. Close the door of your office at home with a sign ‘Do not disturb’. 128(7). 24 8.’ 3. To set them up at home would cost the company $1300 each but it would release $3400 in each office space. Have some basic troubleshooting software on hand.

e-mail and information services different communication and network services required and as there are different applications for teleworking. as well as 347 888 Australians and Canadians. (1995). 1995. Internet World. 14(4). Data Communications. 1994. (1994). The average work time spent in office per teleworker was one day a week. 2. . Over 2 million self-employed professionals do not use any accounting software. pre-existing equipment within the organization. Case 19. Barrett. 16(13). 3. Thus the communication requirements adopted by teleworkers will reflect both the nature of the task (is access to a host computer or LAN required? Is a private e-mail service available? Are voice communications required?) and the prevailing network conditions (the extent of national ISDN service. Self-employed persons with a PC generate almost $70 000 in household incomes 42% more than otherwise. Dec. (1994).A. p. Yarmis. French and Germans. Telecommunications. The computer program that forwarded this Christmas tree greeting was designed to rifle through each recipient’s files in search of automatic routing lists. Burns. The annual savings in facility costs per commuter was $3000 $5000. 27. . 1994. September. p. Personal Computer World. 11(9). Technology. For example. PC Magazine. M. 384 387. D. Nov. PC Magazine. N. Case 19. and Wallace. S. p. Source: Julian Bright. 14. 88 91. Source: Jonathan L. personnel have been told not to execute or store any messages if the sender’s specific purpose is unknown. Electronic mail the state of the art.9 million Americans commute to their jobs either full-time or part-time through their PCs. The average telecommuting equipment was a 386 processor with a fax modem and a dotmatrix printer. 27 30. Open Computing. No provision was included in the program for checking the duplication of names on different mailing lists so the global link was in knots for hours. and Raskin. Telecommunications. Average productivity increases per telecommuter (measured by employers) was 10 16%. the degree of technical complexity used in teleworking applications depends on a number of factors. 102 175. (1988). The average work time increases per teleworker per day was 2 hours. November/December. 85% of telecommuters are equipped to communicate with their employer’s system miles away. and the availability of value added networks. 2. 14(8). Computerworld. The typical system used in home-based systems cost $3000 including software bought at the time of purchase about $1000 more than the average cost of a PC system purchased in the US. according to a spokesman. Caswell. and U. Inc.8: Holiday cheer by electronic mail A Christmas cheer chain-letter sent through IBM’s internal network illustrates the benefits and problems of e-mail. R. R.S. 81. The problem was eventually resolved. and then pass the greeting to each name on that list. (1994). communication costs.2 million Americans communicate through the Internet with 541 949 British. it was observed that: . 45 47. Virtual encounters. The changing face of on-line. T. 229 . The mail must go through. Telecommuting Changes Work Habits. Gasparro. (1995). Feb. C. The result: the card boomerangged through the network.Telecommuters. such as the nature of the work undertaken. Buying e-mail is harder than ever. 22(8). (1993). 22(18). Bibliography Ayre. Engler. Moving LAN e-mail onto the enterprise. News & World Report.7: Telecommuting in the US (in 1994) 3. N. E-mail beyond the LAN. Beard. Further. 103 112. by trapping the files and deleting them. The company was to review operational procedures to prevent future disruptions of its electronic mail service. clogging communication channels so that important business mail was delayed. the cost of the service. 108 175. 1987. Holiday Cheer Brings IBM Net to Knees. Source: Patricia Keefe. or the availability of public e-mail or data services). Teleworking: The Strategic Benefits. 1994.

Secure e-mail cheaply with software encryption. LAN. E-Mail detail.W.Telecommunications and networks Griesmer. 13(8). 18(3). (1994). The Computer Bulletin. Byte. 116 123. 529 552. Roberts.D. U. 111 151. 23(1). Jesmajian. (1994). 6(13). Richardson. S. Computer Technology Review. 7 14. 90 108. (1990). (1993). 230 . Telework: a new way of working and living.B. Who are the high tech homeworkers? Inc. and Wirth. Reinhardt. 1. AT&T Technical Journal. 31 35. 12(1). (1997). Sproull. M. Trowbridge. Telework: Towards the Elusive Office. A. Griffiths. The manager and the information worker of the 1990s. Evolution of messaging standards. 48 50. Smarter e-mail is coming. (1993). L. 2(9). D. 19 30. 10 15. L. 14 17. T. Korte. Information Strategy: The Executive’s Journal. Telework projections. S. (1991). 12(2). Huws. (1993). (1993). Wiley. 14(8). 10(4). Hurwicz. PC Magazine. LAN. R. Maritino. Datamation. Now remote computer users never have to leave home. and Kiesler. M. Runge. Mr. Computers. (1990). (1997). Huws. International Labour Review. Technology. Scientific American. 129(5). P. May/June 21 45. Futures. Teleworking and Telecottages. Strauss. (1991). U. W. 265(3). (1995). 39(23). K. S. and Robinsons. 87 91. R. V. Reichard. 57 62. networks and work.Postman@INTERNET. E-Mail: old meets new. L.

In it. has cyberspace become more mainstream? We shall explore cyberspace through its most ubiquitous manifestation: the Internet. up by 621% over a period of two years. the many uses of the Internet. In 1994. By 1997. the cyberworld where you can find cyberpunks. and the problems of security on the Internet. The common citizenry. We can. there were teenagers that were seduced andranawayreceivinginstructionsontheInternet. He is perhaps right at least in the case of the commercial by IBM in 1995. have little say on how these disputes will be resolved. with its expansion. cybersex. International Herald Tribune. there was obscene and pornographic material on the Internet that promptedtheUSlegislaturetopassabillfiningand sending to jail those who put such objectionable material on the Internet. Cerf In a few years. We then start with an introduction to cyberspace. This we shall do in this chapter.20 INTERNET AND CYBERSPACE Clearly. cyberwork. the potential uses by businesses. cyberphobia. a professor in MIS (Management Information Systems) regularly assigns his students the reading of advertisements in professional magazines. But the courts may have to decide what is and what is not objectionable without violating the freedom of speech that the US constitution guarantees. how can the US government ensure that someone from outside the US will not put objectionable material on the Internet? And so the Internet is very much a political and social issue. It is the fastest growing sector of the economy. We examine the ways to connect to the Internet. 128). Chinese dissidents flashed messages to China on the birthday of the Tiananmen Square massacre that would be banned in China. however. cyberbucks. This cyber interest should raise questions like: Is the cyber-era a hype or a reality? Is cyberspace peopled by gearheads and technojunkies that oppose any intrusion by the commercial world and the government? Or. cybersluts. Vinton G. look at the technological problems involved. Welcome to the cyber-era. Internet had over 60 billion packets transmitted per month starting with zero in 1988. including computer professionals. He argues that commercials are the best predictor of future technology. The Internet is part of the scene of what is called cyberspace.’ Enter cybernuns. there are nuns walking in a monastery in Czechoslovakia when a Mother Superior confesses in a whisper: ‘I’m dying to surf on the Net. and if you chose. Internet is an international net of local networks that connects lots of host computers where information and knowledge resides and can be readily accessed. Even if the courts define what material is objectionable enough not to go on the Internet. 1994 INTRODUCTION At the University of Canterbury in New Zealand. In 1994. There are powerful ‘leveling’ effects to be seen in which elementary scholars and senior statesmen correspond without necessarily being conscious of age and experience distinctions. the Internet and its associated technologies and applications are taking us into new intellectual and business territory. Internet saw many changes: its financing changed hands from an agency of the US government to a firm in the information provider business. it is predicted that some 400 million people in North America will use the Internet at least once a week with world-wide users approaching 80 million (Bayers. there may be more people talking to each other on Internet than on the telephone. 1996: p. The term ‘cyber’ (according to cyberwatchers) has been used in publications at least 1205 times in January 1995. 231 .

. In general. . . . . . populist politics and technology is called cyberculture . One thing that the hackers shared with the counterculture of the ’60s and ’70s’ was a profound distrust of society . transmission by Internet is much faster. . . . By 1989 it had been borrowed by the on-line community to describe not some science-fiction fantasy but today’s increasingly interconnected computer systems especially the millions of computers jacked into the Internet . . . One person who captured this thinking was William Gibson.’ Elmer-Dewitt talks about the shadowy space as: great warehouses and skyscrapers of data . One is to go through a corporate LAN connected through a router and requiring a TCP/IP 232 . encompasses the millions of personal computers connected by modems via the telephone system to commercial on-line services. Still. wires and cables and microwaves are not really cyberspace. . Also. often a few seconds or minutes as compared to hours and sometimes more than a day by fax for large jobs. . it served almost 1 800 000 nodes in some 40 countries. . . It was then supported partially by the US government for its operating costs while the remaining costs were borne by subscribers who paid a fee for connection to local computer hosts that direct traffic to local access providers that tie into the hosts. The term cyberpunk was coined by Gardner Dozois to evoke the combination of punk rock anarchy and high tech. creative human beings . where technology has been the medium for the expression of human creativity . 1995: p. . Eventually. ‘They develop a belief that there’s some kind of actual space behind the screen . Some place that you can’t see but you know is there . basic digital services including e-mail. In this section. Both public and private networks can be part of this loose confederation of networks. documents. . ‘the street will find its own use of technology’ while cyberpunks will use technology to undermine the misuse of technology. 10 11) speed links to local area networks. Internet handles queue (waiting line) control and flow control automatically through its mail protocols. Internet started as a communications service for researchers and academics of the Defense Department in the US. place of unthinkable complexity . we will briefly discuss its evolution and its possible future. . Gibson in Neuromancer (1984) talks about the cyberpeople he knew. Its messages encapsulate fax. This largely unofficial meeting ground of culture. (Evans. . Group information exchange of group news. . the costs of transmission by Internet in 1993 was well below the costs of surface mail and airmail and especially below the $2 costs of a comparable fax. 1994: pp. . In 1993. (Elmer-Dewitt. video as well as many character sets of foreign languages and multiplexed messages across common links.Telecommunications and networks Cyberspace We visited cyberspace implicitly when we discussed Internet and the information highway. mailing lists and other related information are delivered according to group specific profiles to the computers of the subscribers. . with lines of light ranges in the nonspace of the mind. What was restricted to scientific and research communications was soon extended to libraries and many existing databases and accessed by information providers in both the private and the public sector. . . there is a ghost in the machine: the traces of the people who created them . video phone calls. . . who coined the term cyberspace. they cared almost nothing for the forces that moved their employers. office E-Mail systems and the Internet . sound. working within the structure that created and maintained the technology were individual. not the destination: the information superhighway. a science fiction writer. . Cyberspace . . He viewed the free hacker in the future was of ‘Bohemia with computers’: counterculture outlaws surviving on the edge of the information highways . . clusters and constellations of data. banking. . some humanistic values flowed back into the society of hackers . In its past. as well as the millions more with high Connecting to the Internet There are at least three ways to connect to the Internet. 4) Internet A network of networks is the Internet. Internet allows the transmission of mail. They are the means of conveyance. movies and shopping. namely money and power. Internet is expected to provide the combined integrated services of a computer and phone at the cost of a TV today. .

By LAN Internet connection depends on your consumption patterns and what you may have to pay. you have to make a choice.2 Uses of the Internet 233 . There may be some constraints on when and how much you can use the Internet and you may be monitored but for an occasional user this may be no burden. However.1 The Internet connection alternatives connection.Internet and cyberspace Internet Service Provider LAN TCP/IP + SLIP or PPP LAN TCP/IP PC Telephone Line Internet Information Service Provider e. What is the best alternative for you? The answer Local interconnectivity 1. The second route is through a Internet provider and this requires not only a TCP/IP (Transmission Control Protocol/Internet Protocol) connection but also a SLIP (Serial Line Interface Protocol) or a PPP (Point-to-Point Protocol) connection. then a LAN connection may suffice. if you do not have a LAN connection and need access to the Internet then you must explore other alternatives but you must 2. If the organization you are associated with has a LAN (and access to other LANs and the Internet) and you are an occasional user. By Information provider Figure 20.g. By Internet provider Global shopping mall International bulletin board International chat groups International newsgroups International news & e-mail Server sites globally Internet connection Local bulletin board Local chat groups Local newsgroups Local news & e-mail Local shopping mall 3.1. Prodigy and AOL Figure 20. These three alternatives are shown in Figure 20. Prodigy or AOL. The third way is to use the telephone and go through a public information provider such as CompuServe. CompuServe. Given alternatives.

etc.Telecommunications and networks now pay for the connection. For Web. Such free software on the Internet is called freeware (such as the WSGopher configured to serve most Gopher servers) as opposed to shareware that requires you Surfing on the Internet Surfing (working or ‘playing’) on the Internet is not currently an easy or end-user friendly pursuit. This connection will not only give you access to the international shopping malls and international discussion groups.3. and in some cases even sound and video. but it is getting easier by the day. To access and use Web or Gopher. The protocols are OS (operating system) and hardware independent. The WWW is called the Web for short and was initially developed by CERN (European Particle Physics Laboratory) in Switzerland. Access to the Internet is summarized in Figure 20. Transfer Control Protocol. then you need the Internet. In contrast to Web which is hypertext oriented. and even the opportunity of playing computer games. Even the US firms benefit. The Web is the short form of WWW. a GUI (Graphical User Interface) is Mosaic or Cello. Web and Gopher are solutions to publishing information on the Internet. One way may be connecting to an Information Services where e-mail is part of the services offered. where FTP (File Transfer Protocol) allows large files and programs to be transferred between a remote server (computer) to your PC.2. text. but it will give you access to a lot of material that is available on international sites only accessible from the Internet. It was later made into a commercial product by NCSA (National Center for Supercomputer Applications) at the University of Illinois. The Web protocol is a superset of other protocols and embraces multimedia-based systems that are capable of delivering data. This is not too surprising given the fact that much of Internet was developed by the free contribution of time and effort by a large number of people interested in wanting such a system. There is also NetScape (an elegant and powerful interface to Web). systems all around the world. when well designed. However. All the tools and services have a home-page that gives you information of the services offered. One US firm is on record as downloading $64 000 worth of software and documentation in just one year. Archie. Browsing is difficult but easier with excellent software like the Gopher and the Web. as well as any combination of multimedia. Veronica and Jughead (named after comic strip characters) and WAIS (Wide Area Information Services) are software tools used to search archived files residing on FTP sites around the world. allowing Web to be used across different computer 234 . has ‘hot’ links that allows for easy access to related services that are relevant and accessible.. one needs an interface. The Internet connection is offered by some information service providers but for a cost which may be worth it if you have much intentional correspondence and a lot of global traffic downloading (receiving) and/or uploading (sending). The home-page. one of the many newsreader services that is like a large bulletin board specializing in topics where group members can post and reply to messages. There is a lot of freeware and shareware that is both of a technical and general nature on foreign sites especially if you live outside the US. Gopher has a hierarchical structure. A lot of software on the Internet is free. While many Gopher servers are limited to publishing text. The services offered by each alternative Internet connection is summarized in Figure 20. This may be quite adequate and offer other services like news and discussion groups. As a navigating tool for the Internet. Web provides access to servers that are the repository of information and documents (including computer programs) that are of interest to computer scientists. Telnet allows logins to remote Internet servers often anonymously. if the services that you want are abroad. Cello is restricted to Windows systems while Mosaic is open to Windows as well as Macintosh machines and X Windows. voice and images. Usenet also has discussion groups (10 000 of them). There are other tools for the most popular Internet functions. Both solutions are client server based and can link documents on the same server or remote servers. and Usenet News. Web is based on HTTP (HyperText Transport Protocol) with ‘hot links’ to documents through another interface: the HTML (HyperText Markup Language). shopping. World-Wide Web. Mosaic is a very user friendly graphical interface for the Web allowing you to navigate freely by merely clicking the mouse. Downloading or uploading involves a file transfer and a special protocol called the TCP. Web servers can publish text and graphics.

What is needed are standards that allows for an orderly growth and yet maintains the security and privacy of data transmitted. Many of the Internet services are available through commercial on-line services like AOL (America on Line).4 Uses of Internet in bussiness 235 .Internet and cyberspace Mosaic Netscape WWW (Web) Browsers MSN Cello Archie Access the Internet NetScape Tools Gopher Veronica Jughead FTP Telnet WAIS (Wide Area Information Service) Figure 20. one to grab your Internet mail and another to read the postings on the Usenet newsgroups. adding 22 new nations Advertising Marketing to the net making a total of 159 countries). Delphi ComputServe and Prodigy. One of the problems with Internet is that it is an informal organization growing very rapidly (in 1994 it grew by 95%. This is partly being done by standards like the S-HTTP (Secured HTTP) that will secure (largely through encryption) data transmitted to-and-fro on networks and through protocols for security of monetary transactions to be discussed later in this chapter. Home-page Publicity E-mail E-zines USES FTP Cybercash Selling Digicash Gopher E-money Figure 20.3 Access in the Internet to pay a registration fee to the developer. The growth of the Internet is often very ad hoc. You typically would use a combination of software.

there may well be some specification of the issuing authority such as the requirement 236 . Eurocard or an American Express card. In 1994. Various strategies and protocols for security are available but they satisfy only a few of the five desired conditions. where the data on authentication. 3. researchers and the defence agency in the US. Feb. Authentication constitutes ‘what you have’ like a badge. a Citibank card.Telecommunications and networks Internet and businesses The use of the Internet by business was discussed earlier in Chapter 16. but cyberspace is of great interest to business and commerce. Unfortunately. or perimeter-based security. that it be a VISA card. p. For transactions requiring security on networks. where one verifies the two trading parties involved. None of these approaches are inviolate and their presence is no guarantee of the owner since these identifications can be lost. so that it cannot be tampered with and altered. 27. is where the machine (‘where you are’) is secured as distinct from the document-based security where the document of transaction is secured. it is the person making the transaction who is authenticated by what they know (name or password). a demographic population most attractive to businesses that sell to such people. These are: 1. these five conditions do not come neatly packed in a protocol package. Certification of the authenticity of the parties involved. Encryption. badges or cards like credit cards. 5. 2. and even for the communications of e-mail and all documents with confidential or proprietary data. a national registration card or a credit card. The PCMCIA Type II card has a broad array of data elements (including PIN. Confirmation is where there is documentation on the seller receiving the order and the buyer receiving the goods sold. digital signature and proof of ability to pay) Security on the Internet An important consideration for businesses. Studies on the profile of users of Internet show that it is filled with bright. Some methods of verification of identification require special equipment seen mostly in movies and not in everyday business operations. It was not specifically designed for conducting business for a profit. 56 000 businesses were interconnected world-wide (US News & World Report. hand shape. Instead they have smart cards and special cards like the PCMICA card that contain a set of unique information that is difficult to copy or duplicate. stolen or even forged. Other considerations are importants to the Internet and we now examine them. What is more difficult to forge is a combination of biometric features that are naturally unique. 1995. 14). The uses of Internet in business are summarized in Figure 20.4. In this latter approach. The most common are the document-based security measures which includes passwords. Some are more difficult to forge like a driving licence with a photograph but then passports come with photographs and elaborate seals and watermarks (some). certification. Strategies for security can be classified into two main types: a channel-based security. In the latter case. The authentication may be of the ID of each party which will vary with countries but in many countries it is a driving licence. Some of the considerations of security that are applicable to systems other than those using telecommunications are applicable to the Internet as discussed in Chapter 12 earlier. Authentication. voice-print or eye retina scan print). well educated. token or card. non-repudiation and confirmation are all coded. ‘what you are’ such as your personal physical characteristics. A common source of authority would be a governmental agency or the credit card issuing authority. and ‘where you are’ such as the terminal or client identification which can be checked for legitimate access by you. ‘what you know’ like your PIN (Personal Identification Number). there are at least five conditions that must be satisfied. Non-repudiation is the documentation of the agreement between the seller and the buyer specifying the transaction in enough detail so that there is no ambiguity and misinterpretation on the agreement. Internet is of interest to all users of the old ARPANET that was developed by scientist. is security. or what they are or what they possess (some biometric identification like fingerprint. 4. Satisfying all the above five conditions should greatly ensure safety of any transaction though a hundred percent security is never possible. academics. upward mobile people.

However. A random challenge made 3. people outside the US cannot use it. Response to challenge Calculate solution for challenge COMPUTER SYSTEM CHECKING SMART CARD SMART CARD Initiate Checking Process RANDOM CHALLENGE GENERATOR 4a. The US has such good systems of identifying both the sender and the receiver that the US government bars its easy export. Reject card No MATCH? Yes 4b. As international standards develop and these cards follow those standards. process or program only. Shamir and L. This internally calculated solution is compared with the solution of the user. Encryption is a powerful security measure for the Internet and telecommunications in general. The standards for document security now being tested is the SSL.e. Alderman). The costs of such smart cards in 1995 were high. the SHTTP. The system offers a challenge generated randomly which makes the system dynamic and not static where the user can guess. the costs will definitely drop. The steps are as follows: 1. Smart card accepted Figure 20. In the public sector. although they are available abroad. The two basic systems are Kerberps (developed by MIT (Massachusetts Institute of Technology) and RSA (named after the initials of the three founders R. the Secure HyperText Transport Protocol. Request session + ID 2. even though it is based on public key algorithms. 4. This card can be used not only for validating POS (Point-of-Sale) transactions but also for logging on to the Internet through a PCMICA slot in many PCs.5 Security for the smart card that will allow for privacy and non-repudiation protection. drug dealers and mobsters. If there is no match then the card is rejected. A user enters the smart cards in special equipment that initiates the process.Internet and cyberspace USER OF SMART CARD LD + PIN Desire to use smart card 1. which is designed for channel-based security. The process of checking for smart cards is shown in Figure 20. The user then responds to the challenge but can only do so with the knowledge of his/her PIN (Personal Identification Number). map user to specified data. Some smart cards can also recognize the owner. the US has developed the Clipper Chip which codes and decodes messages including e-mail and is protected from snooping by anyone except the government itself. Rivest. It is far ahead of the other standard being developed. 5. then the card is accepted and further transactions are performed. the Securities Sockets Layer. without the risk of an infringement 237 . If there is a match between the two solutions. around £180 each. log user activity and also do privilege mapping. 3. A.5. The system compares this response with a solution that it calculates internally knowing the PIN from the ID. The government claims that it needs a ‘back-door’ key so that it can intercept messages from terrorists. What the world is waiting for is the integrated SSL and SHTTP. i. There are many who cry foul and argue that such measures violate privacy and are merely a way to catch those who avoid paying their taxes. 2.

biometric characteristics Authentication Requirements for secure transactions What you know. e.Telecommunications and networks suit. badge What you are.7 Firewall for security 238 .g. Encryption software in the US is a political tussle between its governmental agencies that want access to data for security reasons and those who consider it a violation of privacy rights.g. e.6 Requirements for secure transactions Telnet FTP Mini E-mail C P U C P U Mainframe Supercomputer Client Server Computer system Restriction Meditation Two-way unrestricted Restricted PC Workstation Web query To corporate file Finger KEY: Figure 20.6 While waiting for internationally recognized standards there are some channel-based security measures. One approach that does control access is the firewall. One is the packet sniffer that is a computer program that runs on a computer site that needs protection and watches all data passing by and records names and passwords that make transactions.g. e. specific client/PC/computer Certification Seller Confirmation from Buyer Non-repudiation RSA Encryption Kerberos Without PIN With PIN Figure 20. key. PIN Where you are. There are many approaches to building What you have. A summary of requirements for security and some of the solutions are shown in Figure 20. The packet sniffer does not stop or even mildly restrict unauthorized intruders but collects data which when analysed can identify the loophole if not the criminal. e.g.

8 Types of security on Internet 239 . it is free (other than the Internet connection). like traffic from a browser seeking information on a server site. In contrast. Some of the types of security on the Internet are summarized in Figure 20.Internet and cyberspace a firewall.8 One approach to security of monetary transactions of the Internet is to use the EDI and EDI VAN. The transactional settlement in the EDI VAN involves a store-and-forward batch processing that is complex. and some traffic is blocked altogether. for example some simple rules on choosing a password.1. One system protects against this by outlawing all common names and even words in a dictionary. Internet EDI Experience: Support: Standards: Security: Reliability: Software costs: Other costs: Response time: Relationship required? Open or closed? Complex? Extensive Good Many Relatively secure Relatively reliable Expensive software Substantial/finite High Yes Closed Yes Internet Little/None None None Not secure No data available Inexpensive software Almost none Very fast (real-time) No Open No that it does not provide instant confirmation. like attempts to enter a sensitive corporate database. costly and slow. One is packet filtering and the other uses gateways to isolate intruders. and it is widely used. quick in transaction confirmation and is the ultimate in impulse shopping. e-mail may go uninterrupted. Firewall Packet sniffer Channel-based protocol (SSL) Credit card PERIMETER(Channel) BASED SECURITY TRANSACTIONBASED Smart card EDI Document security protocol (SHTTP) HYBRID of SSL as SHTTP Figure 20. These and other comparisons of the EDI with the Internet (and the discussion of the EDI in Chapter 17) are summarized in Table 20. One approach is to have a dual set of computers as shown in Figure 20. may be restricted but allowed if the request comes from a known and ‘safe’ site.1 EDI vs. Whether channel-based or document-based. allowing only certain types of access and processing to be done. The Internet is also an open system.7 that funnels all the incoming and outgoing traffic to a system. traffic like FTP (file transfer protocol) or that from finger (a service that provides data on users) or Telnet (a tool of interactive communication) is mediated. the Internet is real-time. The great disadvantage of the EDI VAN is Table 20. One study showed that almost 7% of passwords are names of the user. Some of the traffic. There is also the hybrid approach of combining the filter with the gateway. there are some simple common-sense rules that can be taught to users in training for Internet security.

cautions: . there is still great concern for its state of security. It is self-policed by spamming (jamming angrily) the offender. In one case. It is rabidly democratic. And there is plenty of such motivation around the world. What makes security on the Internet so difficult is that the Internet belongs to no organization or government that can control it. Now the membership is larger than the population of many countries in our world.’ (Wallach. . information was top-down with the editor (of a newspaper or TV programme) deciding on what It can be said of Internet security.Telecommunications and networks Despite the many good things that can be said for Internet security. There is a clash of culture and a difference on how the Internet should be used and what its content should be. Some call it anarchy. there are many weaknesses (perhaps half to three-quarters of the holes in Internet security) that are known to hackers that are not yet openly acknowledged. They are the students armed with a free LAN account number and have time to show their worth. Siegels sued the company for loss of sales and vowed to keep advertising. they can even masquerade as a trusted colleague to convince someone to reveal sensitive personal or business information. All security measures can be thwarted by a person who is determined and clever. Decentralization should be promoted. the greater the challenge. 90 1) Internet. But the case illustrates that there is no organizational entity responsible for what goes on the Internet. as for the security of all information systems. How much cyberporn. 2. 3. 1994: pp. There is a ‘netiquette’ on how the Internet should be used. they so deluged the Siegels couple that the Internet provider cut the Siegels off. Some say that the organization has tenets: 1. there is a new breed claiming membership. These tenets and ethic was easy to maintain and police as long as the membership was a small and homogeneous group dedicated to the Internet. and in providing a community with free communication. Organization of the Internet Security of the Internet is partly due to the nature of its organization structure. For one thing. It is open and non-proprietary. There is often a clash between these ‘newbies’ and the old guard. No body owns the 240 . that no absolute security is possible. All information should be free. The Internet is autonomous (despite some change of financing to an information provider in 1995) and dedicated computer people who volunteer their time and passion. Most of the work is done by dedicated volunteers who pride their work on the Internet. We may not be able to eliminate all intrusions of the Internet but we can try to minimize the risks involved. It crosses national borders and answers to no government. It is still a resolute grass-roots structure. Access to computers should be total and absolutely free. in what they have to display and share. Another commentator on security. cybersluts and cyberculture is healthy and is an expression of the freedom of speech and thought? Why is this the concern of college authorities or even the government? Why is the Internet organized the way it is? The Internet does offer a change in the paradigm of how information is collected. Traditionally. e-mail and other communications can be almost tracelessly forged virtually no one receiving a message over the net can be sure it came from the ostensible sender. . the equivalent of ‘a car left unattended with the engine running. No wonder that Weista Venema of the Eindhoven University of Technology suspects that the most respected domains on the Internet contain computers that are effectively wide open to all comers. when the universities open in the US. And every year. There is not even a master switch to turn it off if so desired. when the NSF in the US withdrew its no-strings-attached financial support and at least one information provider bought into it. The better the security measures. at least till 1995. The motivation may not be money but merely the demonstration of beating the systems. 94). There is no central authority or command to impose a discipline or a standard. 1994: p. or lack of it. The Internet users took the threat seriously and included a Norwegian programmer who has written a program to keep the Siegels off the Internet. Disciplining them and imposing a regime of protocols and control on them is very difficult. It is almost lawless. (Wallach. Electronic impersonators can commit slander or solicit criminal acts in someone else’s name.

9 Evolution of the Internet 241 . 46). 1995: p. The challenge for the citizens of cyberspace as the battles to control the Internet are joined and waged will be to carve out safe. The journalists at the bottom followed the lead of the top and dish out information to the reader. we had shared Internet will have around 100 million servers (projected) 1990s Network Computing (with LAN. weather etc. In the 1970s. play and raise their kids without losing touch with the freewheeling.Internet and cyberspace to include and what to exclude. The magic of the Net is that it thrusts people together in a strange new world. It is a two-way manyto-many relationship. MAN. These differences are summarized in the Table 20. one in which they get to rub shoulders with characters they might otherwise never meet. Table 20. untamable soul that attracted them to the Net in the first place. 2000 Directory of users: Summary and conclusions The Internet can be viewed as an evolution in our modes of computing.2 Comparing information service providers with the Internet INFORMATION SERVICE PROVIDER Audience: Household National Information on news. the flow of information is not unidirectional nor is it a one-to-one relationship. (ElmerDewitt. WAN and the Internet) 1980s Personal Computing (with PCs and workstations) 1970s Shared Computing (time-sharing) Figure 20. For the Internet. pleasant places to work.2. shopping. E-mail Prompt Fixed C Variable Internet PROVIDER Researcher Educationalist International Net of databases E-mail Can be delayed No variable cost Provider is paid Indirectly by governments Planned Services: Delivery: Cost: Control: By business Available in most cases The Internet and information services The Internet is accessed by many information service providers but there are many differences between the two.

We cannot predict the future for the Internet and the confluence of telecommunications and computers for the next ten or twenty years. for example. This evolution is depicted in Figure 20. Under increasing pressure. as well as secure from viruses and outside interference before they become as common or comfortable as a telephone. 10). It is difficult to predict the future of computing and much more difficult to predict the future of the Internet. and browse the Internet server sites instead of using the library. except that the experience will be breathtaking and exciting.Telecommunications and networks computing and time-sharing. At first. That would be an ironic fate for a system designed to enhance communication. Intel agreed to replace all faulty chips. What is more difficult. security and misuse? We do not know the answers to many of these questions. One set of questions relates to publishing. The future role of Internet is very much in the area of speculation. This issue 242 was the subject of conferences in the US in the 1970s and is still being debated in international forums. the manufacturer of the pentium chip. how will the problem of royalties be resolved? There is also the problem of the protection of intellectual property that is now being perhaps addressed by ITO. Or will the Internet be a resource centre for telecommuting and assist the growth and viability of telecommuting? Or will Internet be a channel for marketing and distribution for businesses? How can Internet be managed so that it controls unwanted mail and blatant advertising? How can the Internet be maintained and enhanced. tolerant of human errors. is to anticipate correctly the response of the common person (especially the student) who has yet to accept the Internet. in fact. Intel argued that the flaw would not affect most users and was willing to replace the chip when it determined that this was necessary. Predicting the Internet is a common game among computer scientists as well as futurists. This may involve a generation of change in life-style. in the 1980s. That may. cultural and religious interests? Currently. except to say that computing and telecommunications must be very end-user friendly. prefer corresponding by e-mail instead of using the post office. any more than we could have predicted the consequences of the first telephone or the first computer coming to town. It is difficult to predict the technology and its acceptance in the market-place. These experiences were transmitted rapidly on the Internet including to stockbrokers. Slowly but surely. The human computer interface boundaries will tend to disappear. 1995: p. This is how the world soon got to know about the flaw. who is to define pornography. and sexual prejudice.’ (Elmer-Dewitt. The Internet in this case was the user’s communication device to put pressure on the vendor. informal information can also be transmitted quickly between scientists and researchers as was the original purpose of ARPANET. that some fear that it will ultimately serve to further divide a society that is already splintered by race. Case 20. On the Internet. Internet gateway . it is projected that the Internet may still remain the ‘Greatest Show on Earth’ and have over 100 million servers. in this case Intel. The price of Intel’s stock dropped. and in the 1970s we had network computing and the Internet. politics. and for adults? Can we define allowable content on the Internet that will not offend the many national. Will it remain a large global on-line real-time bulletin board and interactive information service? If so. however. There were delays and much paperwork and formalities involved. Will the Internet go into electronic publishing? If it does. come. the precursor to Internet in 1969. By the end of this decade. the Internet ‘is so fragmented. Internet was not designed for general and extensive use by individuals and business organizations and hence no security measures for such use are in place. robust. let alone the next fifty years. The common person must feel comfortable with cruising and surfing on the Internet and using a PC (or a Internet computer available for $500 in 1996). by going ATM (Asynchronous Transfer Mode) to make it faster and cheaper to use? How could Internet protect against violations of privacy. successor to GATT. It is in this spirit that on 30 October 1994 the mathematician Dr Niecly discovered mistakes in his calculations by the new Intel Pentium chip and checked with colleagues on the Internet. we had personal computing with the ubiquitous PC. who will control the content? Could the content include pornographic material? If there is to be pornography. Or how do you define the broader concept of ‘obscene’ for children and for teenagers.1: Intrusions into Internet Internet is a successor of ARPANET that was designed by the US government to help researchers connect with each other. if not impossible.9.

a British non-religous charity. 14. The rush of letters caused the server through which the firm accessed the Internet to crash and it was eventually kicked off the net. 88. Thomas. an early champion of democracy. ž In 1988. the Lawrence-Livermore National Laboratory in the US. 787. a self-perpetuating worm virus. She comments: ‘Even though I might be logged in on the Internet an average of four hours a day. the ad is unrestricted in size so that one can advertise oneself at any length. ž In 1993.Internet and cyberspace suppliers have maintained that anti-virus scanning is the responsibility of the end-user through security measures are now being implemented and include ‘firewalls’. 1995. The one big problem is that most best matches live on the other side of the world.personal. A good primer for children in cyberspace is: Child Safety on the Information Highway. as well as of freedom of speech and expression. 1995. 9. 23. offers emotional support to the suicidal and despairing through HELP by E-mail service . putting them on par with Suicide hotline ‘The Samaritans. p. Some commentators call the Thomas Web site the first step towards a virtual democracy and cybergovernment. one can get immediate responses.’ Experts advise children (and adults) on the following rules for working on the Internet: never give out any personal or family information. Jan. CERT (Computer Emergency Response Team) at the Carnegie-Mellon University in the US warned network administrators that tens of thousands of Internet passwords had been Others call the Thomas site a rich potential for cyberspace anarchy. Source: Business Week. a mother of two children and a doctoral student at the University of Montreal. Case 20. 2. Thomas is named after Thomas Jefferson. Nov. and never respond to abusive or suggestive messages. Marriage in cyberspace Cybermen and cyberwomen can match themselves for potential marriage by advertising themselves on the Internet by using: alt. there were 1300 ‘incidents’ on the Internet compared with 50 a few years earlier. a law firm in Phoenix. legislators and lobbyists in having equally quick access to future legislation in the US. Thomas site In 1995. in 1993. numerous intrusions into the Internet have taken place: ž In 1986. Source: US News World Report. Jan. found its way into the University of California campus system and wriggled out of control for many days infecting and corrupting thousands of computers on the Internet. p. Meanwhile. electronic tokens and the PersonaCard. . such as numbers or addresses. appropriately called the Internet worm. 1994. a worm on the Internet adversely affected 6000 host computers. . Arizona. Callers are guaranteed absolute confidentiality 243 . The advantages are that it costs nothing. Children in cyberspace There is much concern for children in cyberspace being exposed to pornography and violence. This was the subject of research of Leslie Shade. p. Source: IHT. conceded that an employee had used its computers to distribute pornography on the Internet.2: Bits and bytes from cyberspace Netiquette Canter and Siegel. It was soon ‘flamed’ with letters for breaching the ‘netiquette’ of the Internet. blatantly advertised one of its services on the Internet. No. For a free copy. became the repository of all legislative information available to anyone on the Internet. ž In 1993. ž CERT estimated that. 34. call C1 (800) 843 5678. 60. we would have to actively seek out and deliberately find something offensive. and perhaps most important. a Web site.

The administration sensed the delicacy of the situation because pornographic images were declared obscene by the State Courts just a few months previously. . Slocan also invited questions on e-mail and answered all of them.4 million downloads) and the frequency of retrieval of different types of images including pictures of men and women having sex with animals. Opponents of the university’s action organized over the LAN network a ‘Protest for Freedom in Cyberspace’. It is possible that going on the Internet gave Slocan an image of a leading-edge company and it is also possible that the e-mail helped. and alt. or a combination of the two.’ Whether this control would be at the local level. OAPs. Slocan went on the Internet and issued 244 ž . It is not clear what made the difference but Canfor conceded defeat. The price is charged hourly or half hourly at a rate of about £5 per hour. 1994. e In the US.Telecommunications and networks and retain the right to make their own decisions. Net. pp. Cybercaf´s e The first cybercaf´. the unwaged and the ‘scrounging journos’. control could be imposed from the outside. The study was based on a collection of 917 000 images. It often responses to a request by downloading the necessary program needed. made a hostile bid to takeover Slocan Forest Products. Censorship in cyberspace In October 1994. Feb. 102 4. Source: International Herald Tribune. Customers can even get software revisions through the Internet. e access the home-page on the Internet using the URL: http://www. 10. The administration decided to ‘pull the plug on all major ‘‘sex’’ newsgroups and their subsidiary sections more than 50 groups altogether.arts. Sept. 1995. These cybercaf´s differ in e the food and drink that they offer but the services offered always include time on the Internet. For more information on cybercaf´s. Martin Rimm. In the UK there are discounts and commissions given to students.binaries. Publishers who venture into international networks like the Internet are particularly concerned about libel and slander . the state level.hub. . appeared in Lone don. 21. ž Cyber weapon in takeover war Canfor Corp. 9. No. In May 1995 but they can now be found in cities like Hyderabad in India. a computer systems manufacturer in the US uses its Web site to distribute revisions to its Unix operating system and additions to its printer software.’ The battle lines soon got drawn over the preservation of free speech in the new cyberspace interactive media. The CMU administration determined that the collection in question was part of the Usenet newsgroups. Unless computer users exercise some self-restraint. Slocan fought with the usual defence tactics of newspaper ads and press The core issues were: ‘to what extent can the operators of interactive media be held responsible for the material that moves through their systems? . In addition. it tracked 90 000 packages through its Web site. 16. 10. . detailed arguments as to why shareholders should withhold stock from Canfor. 1995. . A Day in the Life of Cybercaf´. a lumber and pulp company in Canada. Cyberia. with functional titles like alt. 58 63. Nov.’ Source: Internet World. Jan. including the decision to end their life. p.3: Business on the Internet ž Federal Express Corp. Now much of this is done on the Internet being a very costeffective way of communicating with its many customers. Hewlett Packard. Apple Computer makes software updates through its ‘‘home-page’’. The study tracked the usage (6.erotica. 1995. Source: Time. Internet acccess is available (also for a fee) in coffee houses in many US cities. is still a matter to be decided. delivers over 2 million packages each day and had a serious problem of tracking the packages. Case Source: Ed there is no tradition of caf´s like e those in Europe. pp. erotica. a research associate at the Carnegie-Mellon University (CMU) informed the university administration that he was soon going to release his study on on-line pornography.

5: Court in France defied in cyberspace Soon after the death of Fran¸ois Mitterand. was e assured by his lawyers that the ban applied to only the printed version of the book. his personal physician. Dr Gubler asserts that Mitterand was diagnosed with cancer in 1981 and suppressed this information despite the promise to keep the nation informed of his health in national bulletins. Case 20. Meanwhile. 1996. Source: International Herald Tribune. 1995. Jan. Under the new regulations.le. and will insist that providers block pornographic Web sites. while Simon & Schuster has hired a team of computer cybercops to prowl the Internet for new ideas. Plon.6: Internet in Singapore Mr Yao. Christian Hassenfratz. March 6. that provides marketing information on Hershey products as well as corporate information including the history of the popular product ‘Kisses’.com. but that these rights had been placed in doubt by the decision to ban the book. John Wiley & Sons is posting its Journal of Image-Guided Surgery on the Internet. the public prosecutor at Beasan¸on. 4. announced that henceforth all Internet providers and operators must be licensed by the Singapore Broadcasting Authority. which has a Case 20. the publisher of Le Grand Secret decided to take no action because the book was banned. and was soon being accessed at a rate of 1000 calls per hour. 1996. has a 50-page home-page on Web.4: Home-page for Hersheys Hersheys. Time Warner Electronic Publishing is running a serial novel on the Internet with no printed edition planned. 25.Internet and cyberspace ž Well Fargo & Co. 245 . This case raises the important issue of the breaching of book copyrights on the Internet and whether courts have legal jurisdiction in cyberspace for material published in their earthly jurisdiction. exc President of France.hersheys. p. as well e as organizations putting political and religious information on the Internet must now register with the Broadcasting 40 000 copies of the book had been sold and there was a hot market for the book. Some say that this case may be potentially as significant as Gutenberg’s development of movable type. 24. where anyone including the French would have access to the book. 1995. The cost of the home-page for the first year was $4200 composed of the following components: Design @ $12/hour Internet account Ongoing charges $2400 $1200 $12/page Source: Computerworld. all operators from main providers to outlets such as cybercaf´s. The Broadcasting Authority will require service operators to take ‘reasonable measures’ against the broadcast of objectionable material. 7. however. p. He said that if there was any attempt to take legal action against him. of San Francisco gives customers access to their transaction histories as well as their current balance. 27. It was then put on the Internet by a cybercaf´ in Beason¸on. Dr Claude Gubler published a book Le Grand Secret in 1996. 7. available to anyone e c using http://www.web. p. pp. Source: International Herald Tribune. 1. publishers are moving cautiously. and March 19. Case 20. Barbraud says that he had produced the book electronically because ‘the spirit of the Internet is against censorship’. he would immediately transfer the book to a server in the United States. owner of the cybercaf´. the manufacturer of sweets and chocolates. Minister of Information and Arts in Singapore.’ said Yao. acknowledged that Dr Gubler’s intelc lectual property rights had been violated. 70-year copyright protection. Gubler further argues that the President of France was unfit to rule in the final months in power. Pascal Barbraud. The gag order specified a 1000 French franc fine for every copy sold illegally. ‘It’s kind of an anti-pollution measure in cyberspace. By this time. It was used by 2000 customers each day in 1995 who would otherwise have called the customer service representative to discuss their accounts. Nov. Mitterand’s family sought a ban on the publication of the book on grounds that it violated their privacy and that it had contravened the right to medical secrecy. accessible by keying in http://www.

However.’ Source: International Herald Tribune. 13776 Supplement 20. By default. are concerned. The Americans are developing a universal digital code known as Unicode that will allow computers to represent the letters and characters of virtually all the world’s languages. it will be more difficult for them to compete in the information industries of the future. 22% 31% 23% 13% 12% Source: Georgia Institute of Technology. Computer Weekly. Supplement 20. 1994 1685715 550593 113482 81355 19867 7392 6946 Jan. In February 1996. All in all. a group of French researchers put up an all French search engine called Locklace (http://www. 1996. One Korean official states: ‘It’s not only English you have to understand but the American culture. 1995 3372551 1039551 192390 151773 46125 n. p. 1995.’ In France and the Franch speaking part of Canada. 13.S. even slang.iplus.1: Growth of the Internet Year 1971 1974 1976 1983 1984 1986 1987 1989 1991 246 that enables Francophones to find information in any of the thousands of French language sites using French only. English is the language of computing. 27.3: Computers connected to the Internet in 1994 USA UK Germany Canada Australia Japan France 3200000 (rounded) 241191 207717 186722 161166 96632 93041 Source: US News and World Report. They fear ‘that if the language of computers remains English.2: Growth in Internet hosts around the world Region North American Western Europe Pacific Rim Asia Eastern Europe Central & S. and even countries like Malaysia that are very nationalistic about their mother tongue are offering English as a language in order to prepare their citizens for the information age. there are many people who just give up.4: Users of the Internet in 1994 Type of user Education Computer Professional Management Other In Europe 36% 33% 16% 9% 5% In U. people are concerned about cybernauts not being able to use the Internet if they do not know English. March 11. Jan. too.7: English as a lingua franca for computing? The explosive global use of the Internet has raised the question as to which natural language should be used for global computing. Feb. there is some resistance to English as a universal language of computing.a. July 13. American Middle East Source: Internet Society. . 1995. The Japanese. 1992 1993 1994 1995 1000000 2000000 3000000 5000000 Supplement 20. of hosts 23 62 235 500 1000 5000 20000 100000 617000 Supplement 20.Telecommunications and networks Case 20.

247 . 12 Supplement 20. (1995). 1986 The NSF in the US establishes the supercomputing centre which results in an explosion of network interconnections. J. (1995). 4 11. (1996). M. Time. P. Supplement 20. Information retrieval from the Internet: an evaluation of the tools. Bibliography Abernathy. Nielsen. PC World. 13(1). Luotonen. 1995.7: Milestones in the life of Internet 1969 US Department of Defense commissions ARPANET for networking research with its first node at University of California at Los Angeles. 1982 Eunet (European UNIX Network) begins. (1988). A. Ayer. 76 82. and Secret. Duke and UNC. A. Byte.Internet and cyberspace Supplement 20. M. R. Elmer-Dewitt. J. 40 46. Byte. Bryan. Bayers. (1995). 131 146. Internet Research: Electronic Networking Applications and Policy. 20(4). Editorial supplement on The Internetwork Decade. (1995). 1990 ARPANET ceases to exist. 1974 Robert Metcalfe’s Harvard Ph. (1995). (1995). Telecommunications. pp. 70. Baran.D. PC Magazine. Web browsers: the web untangled. p. The State of the Internet. The relative costs in the US in 1995 were as follows: Developing in-house One time costs Software: $500 1000 Hardware: $20 000 40 000 One full-time Web Master: $60 000 120 000 Renting Design and programming: $5000 30 000 Rental fee: $2400 3600 Ongoing costs (annually) Source: Computerworld. Communications of the ACM. A. US. 1984 JUNET (Japan UNIX Network) is established. 1986 Internet worm burrows through Net affecting 6000 hosts. 99 104.5: Build or rent a Web site? One can either develop a Web site in-house or rent one. Wired. Firewalls for sale. 24 27. P. S2 S29. 20(7). Source: CommerceNet. Chapin. N. The results are as follows: Gathering information Collaborating with co-workers Researching the competition Communicating internally Providing customer support Publishing information Buying products information Selling products or services 77% 54% 46% 44% 38% 33% 23% 13% 1976 First use of Usenet establishing connection between two universities. 1994 Mosaic takes the Internet by storm while WWW and Gopher proliferate. and Burke. 1995 Emergence of Java and Intranets. Brinkley. 14(3). (1995). Menlo Park. (1995). 5(3). 4(4). 1995. F. 1994 Shopping malls arrive on the Internet.2 incorporating TCP/IP. named after its football mascot. p. 1974 Vinton Cerf and Bob Kahn detail the TCP for packet network intercommunications.L. Welcome in Cyberspace. and Reichard. 6.6: Users of the Web for business A total of 161 businesses in the US and Canada were interviewed about their uses of the Web for business purposes. 1993 Businesses and media discover the Internet. Bright. Data Communications. R. 29(1). Ellis Horwood. 1991 University of Minnesota introduces Gopher. (1994). 16(1). The Greatest Show on Earth. Spring. 126 128. Time. 37(8). Multiple responses were allowed. Nov. Elmer-Dewitt. 16. 1992 WWW (Web) is released by CERN in Europe. Battle for the Internet. R. The Internet. Spring. K. The World-Wide Web. C. 176 196. Berners-Lee. Oct..C. 69 86. The great Web wipeout. and Computerworld. 3 10. 1983 University at Berkeley releases UNIX 4. Smart Cards. thesis outlines the Ethernet which then became the technology adopted by ARPANET.

Montague.-H. R. Internet resource discovery services. R. IEEE Computer. Datamation. Internetworking: future directions and evolutionary paths. Telecommunications. G. Reichard. (1995). Kosiur.M. Hackman Jr. (1995). 111 137. Wire pirates. Levy. PC Magazine. 27(6). NetUser. 28(1). Scientific American. 38 41. 12(3). D. Wallach. 8 24. (1995). Hurwicz. 14(17). E. September. Postman@Internet. Schmidt. 82 90. L. Oneclick Internet. James. A site of your own. Extending e-mail. and Snyder. Virtual Reality: The Revolution of Computer Generated Artificial World and How it Promises and Threatens to Transform Business and Society. (1995). 49 57. 227 271. P. Computer. 103 7. Vetter. 38 40. 114 125. Reichard. (1995). Internet infrastructure. 2(2). 14 (8). Electronic commerce: Building business security. 174 177. G. (1994). D. S. (1997). Telecommunications. 8(7). H. M. 248 . 33 40. (1997). Krol. March 90 101.R. Fassett. (1991). Internet. Internet. and Montgomery.Telecommunications and networks Evans. (1994). J. K. (1994). K. 7 19. PC Computing. Auerbach. 157 175. Rheingold. 11(3). S. (1992). 29(11). 10 13. Ivine. Mr. PC Magazine. Computing Now!. R. Marion. A. Netscape’s bridge to the Intranet. Summit Books. (1996). 2(1). Issue 3. Danzig. 42(18).J. K. PC Magazine. Man and the Computer. S. Where the hackers meet the rockers. E-money: that’s what I want. J. (1972). Building a corporate Web site: advice for the hesitant. Mosaic and the World-Wide Web. 213 215. Intranets. Obraezka. (1994).P. Wired. O’Reilly. 2(12). 14(8).. (1995). Lipschutz. (1994). Who’s guarding the till at the cyber mall? Datamation. 26(9). The Whole Internet: User’s Guide and Catalogue. (1995). P. 55 74. and Li. A.

These technologies will facilitate the overlay of information and knowledge which will enable corporate managers to pose questions that are fundamental and relevant to them. such as personal computing. Even if the rate of technological growth decreases some. telebanking. the automated factory. predictions of computer technology are always dangerous. . including the development and acceptance of international standards for telecommunications and a rational management of the Internet. International standards for such cellular technology and mobile telephone networks such as the PCS (Personal Communications Services) will be needed. In this chapter we look at the future of telecommunications but only in so far as it is relevant and important to corporate managers. Some will be light and hand-held. We will conclude this final chapter with many predictions of the future with some reflections and caution on why predictions in computing often go wrong and what one may do to minimize its dysfunctional effects if not to eliminate them. office and factory automation. telenews. extract from knowledge understanding and then let understanding ferment into wisdom. Naisbitt in Megatrends 2000 (1984) In a sense. To access such power at all levels of end-users we need better network management. battery-operated (with longer battery life). Our challenge is to process data into information. and other multimedia services. 249 . In integrating such systems across all functions and across all levels of end-users. We need a corporate infrastructure of information that is cross-functional and integrated to provide a careful balance between centralized coordination and decentralized use while at the same time being end-user friendly. In 1993 with seven years to go there are already 12 million subscribers in America alone. And the number will grow as powerful companies merge and consolidate in the US and abroad. As recently as 1984. refine information into knowledge. The richness of such a society will also depend on advanced applications now under development such as the digital library. But some computers will be wire-less which will mean that they may face the vested interests of companies that have investments in over 200 million lines of copper wire (in America alone) currently used for transmission. we will use our brain power to create instead of our physical power and the technology of the day will extend and enhance our mental ability. virtual reality. Turing (1950) We can expect changes and improvements in computing technology. and networking. The smaller end of PCs will run at supercomputer speeds and will be simple to use. we may approach a telematic society. and capable of communicating across the land and even across the world with digital microcellular technology. the electronic office. However. These are essential to applications like home-shopping. The problem will be to find an IT platform that allows the convergence of the critical technologies in IS. we have automated the process of gathering information without enhancing our ability to absorb its meaning . the use of words and general educated opinion will have changed so much that one is able to speak of ‘machines thinking’ without expecting to be contradicted. there will still be faster and more cost-effective computers. One concern is the growth and trends in the increasing power of computers and telecommunications. etc. AT&T predicted that there will be 900 000 wireless telephones by year 2000. . Al Gore (1991) At the end of the century.21 WHAT LIES AHEAD? As we move from an industrial to an information society. A.

With the use of a centralized supercomputer the costs are expected to be lower still. 78). booms and data-gloves for achieving virtual reality. George Gilder. (Gilder. and so on. hand recognition devices) and interactive 3D graphics. cable and TV. consider that in 1994 there were on-going feasibility studies of storing thousands of pages of text and thousands of films (full feature films not 10 minute documentaries) under the control of supercomputers and accessible to a homeowner. We may well see the merging of some telephone. . Teleprocessing has made control by remote means possible. Locationless conferences conferences without meeting places. . . Telecommunications will also see great strides in performance and the law of microcosm will soon merge with the law of telecosm. libraries. 250 . This advanced communication environment will not only facilitate applications like videoconferencing. One such prediction is that we will soon have a three-giga machine (where giga is a measure in billions): giga-instructions per second. With the use of fibre optics. This will reshape our current distribution channels and will give access to the end-user that will be not just quick and easy but also comfortable. Currently. As a preview of such a world. in games rooms. there will be computers used in every place imaginable: in automobiles. expanding its capabilities and potential bandwidth. It would also enable salaries to be closely correlated to performance. . gigabyte bus and gigabytes of memory. selected and manipulated at will. the more efficient and powerful are its parts. when combined with end-to-end digitization. Locationless management management without headquarters (von Simson. medical consultation.g. and public knowledge-bases where knowledge can be interacted. the proportion of computers in America connected in networks rose from below 10% to over 60%. the law of telecosm finds the same kind of exponential gains in linking computers: connect any number of n computers and their total value rises in proportion to n2 . Between 1989 and 1993. teleshopping and teleconferencing. 1993: p. in a peer-to-peer computer arrangement. in aeroplanes (already so in the Boeing 777). There will be more voice processing. may result in what Professor Solomon of MIT predicts: ‘the public switched network will be transformed into one large processor’. Locationless selling selling that requires no retail store. states this well: Just as the law of microcosm essentially showed that linking any number n of transistors on a single chip leads to n 2 gains in computer efficiency. This may be good for the employer but not necessarily for the employee because of the danger of personal privacy being violated. Locationless inventory inventory that needs no warehouse. offices. libraries. joint ventures and alliances.Telecommunications and networks The future in the context of the past Before making any predictions one should remember the caution of Niels Bohr. Locationless training training without a classroom. This could have the advantage of providing more information to personnel about their own performance and be used as a coaching device. in parallel with the development of an electronic information highway and a NII (National Information Infrastructure). The computer paradigm with its high bandwidth in both fibre and the air may replace the telephone. We will also see the use of helmets. 1993: p. but will also provide access to knowledge on entertainment. will give us access to anything in computer storage and do so interactively. Such merging. goggles. 106) Locationless implies interconnectivity of computers by networks which will continue to grow . the eminent Danish physicist: It is hard to predict especially the future. factories. each new device is a resource for the system. There may well be need for legislation to regulate monitoring. cable. The laws of microcosm and telecosm. TV and computer companies in alliances with firms in the publishing and entertainment industry. There may even be many computers that will be locationless. And distribution will be much cheaper. the share of the entertainment dollar going to distribution may be five cents. This capability in conjunction with visual processing has greatly increased the possibilities of monitoring personnel. doctor’s offices and waiting rooms. But there are some things that one can predict with some confidence. which will enable us to leave the computer screen and use our body to interact with a rich variety of virtual objects to replace the physical world with a computer generated one. some 30 cents of every dollar in the US goes to distribution. With mass ownership of computers and with interconnected computers. biometric input (e. The larger the network grows.

telemedicine. however. This seamless flow should not only be nation-wide through a national highway infrastructure (also called the superhighway or the Infobahn) but also globally through a GII. but instead allow for a network system that is efficient and effective. digital libraries. and future. and for the inclusion of business traffic on the GII. to restructure which may have to be a shift from a vertically integrated industry.What lies ahead? Trends in telecommunications technology In the 1990s. This can be achieved partly by integrating existing islands of computerization. who may well be vertically integrated. What is needed is the movement towards systems that have an open architecture and are digital. integrated and enduser friendly. This distribution will be greatly facilitated by advances in the implementation of the top layers of the ISO/OSI architecture and by digitizing information. present and future. There is a great overlap between developments of the past. The transformation of computing by telecommunications and networks is summarized in Figure 21. telenews and telecommuting. to the present and to the future. it would be necessary for many countries (especially the large trading partners) to change their national laws as they relate to the regulatory environment (for example. and enables a seamless flow of network traffic.1 is very approximate. text. Such a highway infrastructure in some countries (like the US) will not be owned and maintained by the government but developed. do not unduly restrict consumer choice and flexibility. Hence. low telecommunication costs and better privacy protection) and change their labour laws (such as those allowing for flexible time and telecommuting). There is a long gestation period for telecommunications technology.1 Trends from past present to present future 251 . we translate information (data. processing and distribution of information. We need systems that support the possibility of applications including telebanking. It shows trends from the past. the time horizon LAN/MAN/WAN APPANET/Internet Proprietory systems/ Platforms/protocols/ Objects for Figure 21. audio and video) into 0’s and 1’s to facilitate the efficient storage. It may be necessary. in which a single entity provided LAN/MAN/WAN/GAN Info. present. Global International Infrastructure. We shall see greater use of ISDN (Integrated Services Digital Network) and enhancements of ISDN such as B-ISDN (which uses a fixed cell size with asynchronous transfer mode cell switching technology). yet can be quite revealing. Also needed is that the vendor providers. superhighway/Internet OPEN systems/ Platforms/protocols/ Objects Digital world Broad bandwidth (Giga bits) Video communications Multimedia End-User friendly systems WIRED CITIES TÉLÉMATIQUE SOCIETY Interconnectivity STANDARDIZATION Analogue world ISDN Narrow bandwidth ATM User unfriendly Human factors & Ergonomics systems FUNCTIONAL APPLICATIONS INTEGRATION Figure 21. It is difficult to slice time into neat dimensions of past. telecommunications will continue to be concerned with the distribution of information (and knowledge). broadband. For this evolution from ARPANET and through Internet. The transmission will be over high speed electronic highways. owned and managed by the private sector. In digitizing.1.

The private networks have the added advantage of strategic control over their operations. . (The private networks do not always belong to one firm but could include hardware manufacturers.2. Outside the US. again to control) what is variously known as the ‘information superhighway’ or ‘infrastructure’ . . . private networks using the Internet. Restructured and integrated or not. . forces essentially every player to harmonize its products and services with those of other players in other layers at least to the extent of assuring that products and services from the different layers can work together. 1993: p. There are other support technologies that are not new and have been discussed earlier but they will appear in the future with robust enhancements and greater functionality. Apart from imposing a few familiar safeguards. applets (programs with specialized functions in Java). software houses suppliers and even public carriers). We will also see a restructuring of the telecommunications industry. Traditionally. the changes are neither a big threat to culture or decency.1. telecommunications has been a public company. Given the history of strong PT&Ts there is resistance to privatize by governments who are fearful of losing control. The new diversity of telecommunications markets . A summary of the comparison of private and public networks is summarized in Table 21. horizontal structure . . systems could use supporting technologies. These are all listed in Figure 21. to a richly diverse. 1995: p. 13) Digital Wire-less Applets Knowbots PDA INTELLIGENT SYSTEMS Java P C S Bandwidth-on-demand ATM backbone VIRTUAL SONET SWITCHED 56 SMDS Software agents Intelligent Hubs Intranets Figure 21. (Heilmeier. many national PT&T (Post Telephone and Telegraph) companies are being privatized. or culture. (Economist. and knowbots (robots working on a knowledge-base). cable and computer industries are no longer forbidden to compete on each other’s territory. the cleverest thing that governments can do about all these changes is to stand back and let them happen. when active will perform specific tasks. intranets. 31). and they have unique features and services to offer. . Others will say that it is the job of governments to build and shape (and so. Feb. 25. Customers are not locked to a relationship established with a single provider of goods and services . These support technologies for telecommunications include software intelligent agents (agents are programs that when passive will monitor. they are faster to implement because of the lack of government bureaucracy found in the public sector. In the US. . . but because of its high cost and low responsiveness of technology and consumer demand. . TV.2 Supporting technologies 252 . firms started their own private networks. the telephone.Telecommunications and networks not only a telecommunications service but everything behind it in a network and on the customer’s premises. and when a master agent will customize applications like summarizing information from a newspaper). The fearful ones will try to impose the dead hand of regulation (in the name of protecting privacy. . or of clamping down on pornography or crime). nor a panacea for jobs .

Both hope to have one device for both the telephone and TV. We do not show the fourth dimension because of the difficulty in displaying it on the twodimensional space of this book. All this is just one dimension. or a combination of these. with Germany not far behind. They are all dashing around trying to capture the market in as many cells of the market as possible. one for text and e-mail and one TV pictures and films. and France still far behind. This results in intersections of interests and so there are strange alliances between some firms that are in joint ventures in one market but in fierce competition in another. but video companies and book publishers. or else go it alone Uniquely configured for organization By government agency Implementation: Slow to adapt and learn Standards: Tendency to wait for standards Generalized Features There is a strong trend towards deregulation of telecommunications in industrialized countries. MAN. 253 . The market though is too large for any one firm (even titans in large industries) and so the firms are prancing around making alliances and mergers. or both. or distributing it. telephone. WAN or a global network. or satellite. The value-added information related industries will soon be open to competition though there is still a strong feeling in many countries that their telephone and computer industries are too important to be left to the private sector. voice.What lies ahead? Table 21. video and multimedia. There is actually a fourth dimension: content. The dimension of content brings other large oligopolies into play that include not just film companies in Hollywood. (France supported its ailing computer firm with 10 billion French francs in 1992 93).1 Public versus private networks Public Cost: Funding: Quality: Services: Security: Control: Higher Public High Generalized Good to high Private Lower By firm Adequate Customized Adequate to high Strategic control Quick to adapt. And there is also a third dimension: that of media which includes data. learn and innovate Consider if any. graphics.3. Each wants to capture the market where it has a comparative advantage which may be a row or column in our matrix of Figure 21. The two main players in this game are the telephone and the cable companies. Each firm may have a set of suppliers where some are multimillion dollar firm themselves. We see deregulation of PT&T coming in Europe too though at varying rates among its telecommunications receptive nations. shared and even intelligent networks in the future. Each cell in this three-dimensional matrix is represented by one or more company. This matrix is shown in Figure 21. Some technical experts think that this may not be possible because a TV that can produce a quality picture may not be able to produce good quality text which has the greater demand. cable.3. The firms involved use different media like fibre optics. What may happen is that we shall have two devices. The UK is very committed. The US broke up the large monopoly of AT&T in telecommunications and allowed all communications companies (telephone and cable) as well as computer companies to compete with each other. TV. These companies that have the content want an alliance with a carrier to carry their content just as much as carriers are looking around for content to carry. It is generally agreed that the high rate of innovation and integration of such companies is largely due to their deregulation. Privatization has added more alternatives to public. These firms are not small dwarfs but are often giants of industry that will invest billions in just testing the market (like Bell Atlantic that invested $11 billion into testing a fibre optic and two-way TV system in 11 million homes by the year 2000). We all know about the bigness of Hollywood studios but there is less visibility of the many video companies and publishing houses that are large and powerful. Their strategists are debating whether the most profits are in owning the content. The other dimension is the scope of the network: a LAN. some companies encompassing more than one cell. value added. Some of these media can be offered in one place including an access to the Internet instead of separately through a combination of hardware and software which makes hardware manufacturers and software houses (some being billion dollar companies) important players in the market.

Telecommunications and networks MEDIA Multimedia Video Graphics Voice Data IMPLEMENTATION LAN MAN WAN SCOPE GLOBAL Public Private Other Figure 21. offering security and protection against invasion of privacy.3 Dimensions of networks This new mix of competition and cooperation is so important that it has been given a special name: coopetion. which in conjunction with services such as the Internet will help 254 us towards a global information network and a t´l´matique society. The problem of protecting intellectual property rights globally also is especially difficult when the traditional paradigm of author publisher library may now be disrupted. In 1996. ee Meanwhile. ‘The desktop computer industry is dead. long distance and wire-less) to a population of over 50 million in the US alone. Steve Jobs. one founder of the Apple PC and later the father of the NeXT computer. In this battle for the market which is large and growing at a prodigious rate there may well be a shakeout with the strongest alliances surviving. establishing standards that are globally acceptable. for soon it may be cheaper to store a video in computer memory and deliver on demand and store it on magnetic tape or optical disk. There are also the cautions of authors like Postman in Technopology about the surrender of culture to technology like the Internet. both Internet and any telecommunications infrastructure face common problems of defining and then providing universal access. controlling ‘obscene’ material. like the video retailing industry. correctly predicted in the early days of the PC that a PC will soon be on every office desktop. and metering users and usage so that any charging system is fair and equitable. an on-line computer service galaxy. The growth in demand for information services will also increase because of an information superhighway and NII (National Information Infrastructure).’ He then predicted that ‘people are going to stop going to a lot of stores. They may then get a foothold into what may soon (around the turn of the century) become the Infobahn or superhighway of the future offering a combination of services for different distances (local. Some firms and even industries may have to exit. This will facilitate our approaching cyberspace. Postman cautions that we may soon have a diverse and rich set of information without knowing how to handle it. The liberalization of the information industry and a change in attitudes in some countries toward freedom of transfer of information across national borders (such as in the Eastern European countries that are no longer tied to Russia) will greatly increase the growth of demand for information and knowledge and their related services. Jobs asserted. He has the fear that we well lose our community responsibility and instead rely heavily on chat sessions and downloaded information from the Internet and the Web. many most likely being oligopolies. And they’re going to buy .

and so it is having financial problems. planned for initiation on 1 January 1998. in France. Another demand-driven development is the trend towards a GAN. By the year 2000. the dismantling of the LAN. Each net pays its own share with the backbones being supported on a national or regional basis. in Geneva. 1996: pp. designed by the US Department of Defense as a communications system for the cold war. If the Web got up to 10 per cent of the goods and services in the country. In addition. However.’ (Jobs. its sources of income are drying out and the Internet needs reform and new management. There are technological problems like insufficient bandwidth but improvement in performance of transmission media and component technologies will soon overcome the bandwidth scare. in the US. there are minis and mainframes that use networks for daily transactions like reservations. and the supply side. It was never designed for commercial use and is inadequate for the business and commercial needs of our modern world. it would be phenomenal. Many of them will be using Internet. . the increasing globalization of our economies. cable. TV. but I don’t believe we live in an information society. there were over 60 million users of Microsoft Windows that could use networks for electronic shopping through credit cards and access information from public on-line information services like CompuServe and Prodigy which in 1994 had over 2 million subscribers each. We live in an information economy. Its early development was the ARPANET. MAN and WAN. it is CERN. satellite and wire-less cellular phones) are now being opened to competition by being auctioned rather than being licensed by the government. Global Area Network. Eventually. This does not mean Management of the Internet An important and urgent question for network management concerns the Internet. . 103. Switzerland. On the way to a GAN. electronic publishing. . There is a great need for open systems in hardware and software components of networks allowing them to be interchangeable and interoperable. Thus. Furthermore. financial services. there is the counterargument that Internet is too ad hoc and informal. . that is. and the increase in multinational corporations and multinational cooperation between national governments and regions. 104 5). I think it’ll go much higher than that. The five channels of communication (telephone. This will be a response to the need for a better architecture suited to commercial operations. it is estimated that there will be over 100 million microcomputers tied into corporate networks. though through a different route: the Open Network Provision. but rather their expansion and extension. and for the 18 European countries. . the probable availability of new technology relevant to network management. the Corporation for Research and Educational Networking. In 1994. We’re already in information overload . which is a confederation of networks with each having its own opinion on how things should work. The desire for competition and the ‘levelling of the playing field’ for all players in telecommunications is also being pursued in the European Union. it will become a huge part of the economy . The Internet has matured. This then allows plug-and-play computing with products from different international vendors to be plugged into different platforms of integrated systems. and other commercial applications including interactive ones. Since the cold war has cooled off (and because much of the network 255 . . The Web’s not going to capture everybody. the demand for more network services. the National Science Foundation was the payer.What lies ahead? stuff all over the Web . it is the EASInet partly funded by IBM. Network management and the future There is much that network management must respond to in terms of the changing environment of telecommunications and networks. It is the outgrowth of ARPANET that was designed in the US to allow researchers and educationists to ‘talk’ to each other electronically. one will perhaps see a different architecture. We look at these developments from two points of view: the demand side. Networks are growing continuously and at an explosive rate. EFT (Electronic Fund Transfer). It has been argued that we already have a GAN in the form of Internet. The problems of the future are not technological but organizational and financial. more advanced network technology and a national infrastructure like the telecommunications superhighway (or information highway) in the US.

say.Telecommunications and networks architecture had stabilized). The fee is also based on the bandwidth and so it is lower for a home user of 9 600 bps 256 as opposed to a 56 kbps by business users. . The nature of demand is also posing unique problems. if so. ‘User A sends a 100 byte request to User B. telecommuters and others working at home paid a fixed fee. these organizations (like information service providers) may demand a say on how the Internet is to be organized and who pays what. The information can be collected and protocols adapted but the cost of doing so will require passing the extra cost to the customer and may not be worth the trouble. regional and backbone services as well as the access providers. free e-mail and downloading from all round the world. usually a fixed fee plus a variable fee for the time beyond the ‘free’ time. the Internet was not free for everyone in recent years. But this can raise unique problems on the Internet. A recent increase in the price of the Internet usage in Australia resulted in a sharp decline of usage by keen students. The price elasticity of demand is expected across a broad spectrum of users though hard statistics are hard to come by. even though User A initiated the transaction and received all the benefits. a doctor wanting a CAT scan of a patient who is dying. There are other schedules of fee and this emphasizes the point that some income is collected by fees. the Defense Department withheld its support for network development but the National Science Foundation (NSF) continued to support operational expenses in order to help the Internet build its market. 29). the NSF said: ‘the market can stand on its own without our seed money. The Internet is so successful that recently there was talk of the danger of a meltdown.’ and withdrew its financial support. Melting of the Internet could occur once monetary transactions on the Internet are secure and businesses start using it not just for advertising and marketing but for operational traffic.’ (Brody. Another approach is to keep the customer fee nominal but to charge businesses for their advertisements and for setting up shop in the cybermall. how much should they have to pay? It has been said that the Internet is autonomous and self-policing. This approach is very inequitable. And so there are no longer any ‘free lunches’. Can such messages barrel down the Internet with the highest priority much as ambulances have priority on our roads? And how about a business wanting an important teleconferencing session but is behind a massive downloading of a computer program or a film? Who has priority? Should businesses get priority if they have paid for it? And. Then. That is largely true but there is a governing body of representatives from government as well as educational and research institutions that determine prices that can be charged and standards that must be followed besides the informal netiquette. and more so if the charges would depend not just on the size of the message but also on the distance travelled. The TCP/IP protocol used on the Internet was developed for the ARPANET switching ‘packets’ of data and has no way of providing the detailed information on transactions needed for a charging system. The actual prices will have to be negotiated with the providers of local. Anyway. the withdrawal of support by the NSF will hurt the non-profit-making organizations and perhaps even academia. A na¨ve billing sysı tem would charge User B 10 000 times more than User A. 1995: p. This would dissuade users from getting what they want from wherever it is best available. How can we reduce the burden and pain for users without dampening their enthusiasm and bona fide use? Will pricing more on the Internet make it unaffordable and thereby destroy the diversity and populous nature of the Internet? How can we protect the freewheeling exploration and experimentation that takes place in the on-line society that the Internet is? How can we let researchers and libraries searching for information get what they want and yet control excesses and misuse? What is the value of public discourse and community services as opposed to commercial opportunity on the Internet? How can we prevent businesses from taking advantage of the system without limiting their right to make legitimate profits on the Internet? How can we prevent large businesses from swamping smaller ones? Pricing is crucial to usage of the Internet. who responds by transmitting a 1 megabyte program. Herb Brody gives the following example. The use of telemedicine raises the problem of assigning priorities to real-time processing by. The Internet has to be selfsupporting. Actually. They may have to assign priorities and fix prices. It almost occurred after the bubble telescope sent tons of data and swamped many users on the Internet. And they may have the company of others as organizations start subsidizing the Internet and replace the NSF benefactor. One solution to pricing may be to charge for each unit consumed just as do the utilities. However. this may not be financially feasible. in 1995. Businesses.

We may have a similar problem with the network systems architecture. Interactive communication tools will stimulate empathy with others and help bind human ties. The Japanese have not planned fibre to the curbs of homes. The questions we must also ask is: Will the quality of life in a t´l´matique society be improved? ee Optimists see freedom from drudgery. intelligent management of natural resources. and interactive TV and movies. A t´l´matique society ee Telecommunications provides the glue to integrated applications especially for dispersed computing whether this be internationally or locally. In Europe. security infrastructure. they say. the absence of standards leads to uncertainty in the market which breaks the cycle of innovation and creative productivity. and messaging management. the X. You do not want to standardize too early in a fast moving technology like telecommunications or else you run the risk of freezing technology. on-line databases. Commerce and Transport. That is not good for the consumer nor for world trade. homeshopping (teleshopping).What lies ahead? A variation of this approach is to tax each business and use the tax money for operations and to subsidize libraries and community centres. the EDI for the Administration. At the city level it would include e-mail. Also. We have the international X. asynchronous access. and the elimination of war and poverty in a t´l´matique ee society. the Control of the Twenty-First Century. The danger of course is that there will be no merging and common standardization of messaging architectures and that the two systems will continue to develop along separate and parallel paths. . ee It is the integration of telecommunications and computers that led to the term t´l´matique ee society (telematic society. The hierarchy of applications that could lead to a t´l´matique society through wired ee cities is summarized in Figure 21. They predict a new Renaissance since more time will be available for leisure and cultural pursuits. At the regional level it would include the Technopolis Strategy in Japan where a number of wired cities (technological metropolises) are connected by fibre optics. once we have many wired countries and technopolis integrated with each other. . Hopefully these two standards will converge over time on important considerations of addressing and naming. In the US. High Technology. The home will become the focus of daily life. the electronic newspaper. A glimpse of the future is described in the book Technopolis Strategy (Tatsuno. but not in the US where the Internet standards are in vogue and fashion. At the same time you need international standards accepted by the developed countries or else vendors around the world with supporting applications will be inhibited less their product does not sell and becomes obsolete. At the world level. The problem of international standards harmonization is a difficult one.400 adopted by many national governments in both public as well as private sectors around the world. One example would be the differences in standards for EDI. Europe and Japan and many countries have adopted the ISO model of OSI but the US and the large vendors of telecommunications systems support SNA by IBM and the TCP/IP used by the Internet. promoting family togetherness 257 Standards Another of the uncertainties of the future in network management concern standards. we approach a t´l´matique society. electronic home. Access to the world’s knowledge will contribute to mankind’s understanding of the future.4. The problems of a t´l´matique society is less ee technological and more social and political. a concept originally coined by the French administrators Nora and Minc. The subtitle of this book is worth noting: Japan. Another example would be the existence of two parallel standards in messaging. there is the EDIFACT. They were asked by their President to assess the potential of telecommunications and computers and the danger of the American monopoly affecting their culture and turning it into a ‘McDonald’s society’ . home banking. the electronic office and the automated factory. Some of the world-wide integration may be in the far future but integration and telecommunication make them a viable possibility. Such integration of electronic homes into wired cities and then into regional integration is currently the approach taken by the US. in English). At the corporate level this would include . The technopolis strategy in Japan is to integrate regionally with 65% being fibre used by corporations and intercity communications. 1986). A national ‘Next Generation Communications Infrastructure’ is planned for 2015 at a cost of $410 billion.23 developed by ANSI (American National Standards Institute) is in vogue.

will sever all ties with human. ISDN in a historical perspective may be revolutionary in the sense that it will transform an analogue world of telephony to a digital world of computers and telecommunications. They see the responsibility for managing cities. He sees: A civilization no longer required to put its best energies into marketization.4 Hierarchy for a t´l´matique society ee and family values. with imaginary constructions of the world rather than the physical realities of every day life. but more enhanced versions. people will filter out unwelcome news. waited on hand and foot by domestic robots. a ‘practopia’ that is neither the best nor the worst of all possible worlds. to chose a single example and inventing new ethical or moral standards to deal with such complex issues. in favour of computer companionship. 258 . For example. 1989: p. advances and innovations in information technology itself will not decide our future. There will be more of the same in many network technologies. running factories. National boundaries will lose importance as the world community is born. finally. preferring to live with virtual realities that is. (Toffler. A civilization capable of directing great passion into art. We have a destiny to create in the t´l´matique society. Pessimists see illiterates glued to game shows on video screens in the world of tomorrow. Clearly. ee A t´l´matique society. Surveillance systems with little regard for personal privacy will be the norm. In the Third World. with implementations in the late 1980s and early 1990s. and distributing goods and services delegated to intelligent computers. from the view point ee of technology. they say. We must begin now to construct a society that can meet the revolutionary changes of advanced technology. It is the way technology is used that will decide the quality of life in the years ahead. 375). A civilization. A civilization facing unprecedented historical choices about genetics and evolution.Telecommunications and networks Télématique (Telematic) Society Regional integration Regional integration (technopolis) Wired city Typical wired city Wired city Wired city Electronic home Automated factory Teleshopping Telebanking E-mail Electronic office Interactive TV On-line Information Services On-line databases Figure 21. that is at least potentially democratic and humane. in better balance with the biosphere and no longer dangerously dependent on exploitive subsidies from the rest of the world. represents no great revolutionary trends on the immediate time horizon. Individuals. In the future we will see advanced versions of both ATM and ISDN and likely in the open system ‘plug and play’ mode. we have heard of ATM (Asynchronous Transfer Mode) and ISDN (Integrated Services Digital Network) throughout the 1980s. With self-selective media. which will indulge every individual selfish wish. Alvin Toffler sees a middle course for society. growing crops.

Collecting royalties will be a problem if the collection has to be done across national borders and the receiving nation is not willing to pay royalties.9 billion voice messages in 1994. ROYALTY MANAGEMENT Publishing House Advanced applications Many of the applications shown in Figure 21. which is not new in the conceptual sense but new in that it has never been fully implemented. These may include the digital library. Adaptation of new technology often takes long. more users of network computing and more integrated systems. Even in 1996. Standardization will also help in the shift of a proprietary platform to a standard platform and proprietary protocols and objects to standard protocols and objects. Witness the case of the fax: the first fax message was sent from Lyon to Paris. Such multimedia will improve applications such as those using 3D graphics in CAD (Computer Aided Design). medical applications. It took over Manuscript sent electronically (FTP) AUTHOR Rejection notice Existing documents 6 SCANNER 4 COPYRIGHT Access/Queries/Requests READER Responses to queries and requests for materials DIGITAL PROPOSITORY & COPYING FACILITY Figure 21. This is to be expected and is related to the problems of copyright management. Standardization will also help in more and better applications. there were over 10 million fax messages delivered. This is part of the shift from proprietary to open systems. This was in addition to the 11.4 will survive in the future and may well be enhanced. which in the mid-1990s was anywhere between 6 and 18 months. This involves not just assessing royalties. but collecting them and distributing them. Greater use of ATM and ATMLAN switches will change the narrowband world to a broadband world (with gigabits per second). (During 1987 94.5). have wide-areas connections across the enterprise. reduce delays in telecommunications. The review management process in the publishing house may not change much though the roles of the author. 125 years for it to be universally accepted. once a manuscript is accepted. It shows that the author no longer sends the publisher a typed manuscript but sends it electronically (Path 1 2 in Figure 21.What lies ahead? The replacement of telephones will be accelerated by the ATM.) Back to the digital library. the publisher. Instead. the librarian and the marketing department of a publisher will change. the publishable material can be available within days (Path 2 4) greatly compressing the time now required for publishing (including a possible integration of multimedia material) and distribution. in the US alone. more LANs. One view of the organization of a digital library is shown in Figure 21. There will be other applications that are innovative and advanced. the US had problems with the acknowledgement of copyrights 2 Review Editors/Reviewers Management 3 No OK? Yes 5 Royalty accounting info. The process of publishing will also change and will no longer be a mechanical process. and increase applications such as interactive video and multimedia.5 A view of a digital library 259 .5). The publishing house will be spared much of the publication headaches but will have new ones such as royalty management (Box 5 in Figure 21.5. video-conferencing and entertainment.

Reading off a screen is not as good as reading from a printed page. Video monitors will replace blackboards. teachers and reference librarians will have access through the Internet to materials and people (through chat sessions) around the world. The reader may wish to project the pages of the book from a digital TV on the wall (or ceiling) and read while lying in bed! What if the reader wants a 600 page book and does not have a fast laser printer at home? That reader may well have to go to the local distribution centre that may replace what was once a local library. networks. It may well have to be decided through international treaties or through international organizations like the WTO (World Trade Organization). workers Teleworker Figure 21.6 Consumers. has flickering that may be uncomfortable or injurious to the eyesight. media and services 260 . These alternative for the representation of material are relevant not only to the person at home but also to the student in school. This will be no problem for future publications because most of the input of new materials will be in machine readable form. Such access will also ‘level the playing field’ between the rich urban students and the poorer rural population offering equal access to all educational resources (materials and teachers). The problem is not technological but legal and political. This empowers the student with control over what can be accessed.Telecommunications and networks and the collection of royalties for software and music created by US citizens. Many of the media currently available. Or may be there is a flat screen display that is lighter than a heavy book and can be read while lying in bed. and certainty poses constraints of location and posture of reader. Also. Other service (and media) not listed may well emerge through the creativity of the NETWORKS & MEDIA NETWORKS LAN MAN WAN MEDIA Computer Televisions Digital TV Cellular phone Digital notebook Video-on-demand Movies-on-demand E-mail and fax Hypermedia Collaborative work Intelligent agents SERVICES Educational institutions Museum access Digital library Religious groups On-line information Internet Electronic shopping Electronic news Financial Medical Other (P) Consumers Home PC user Businesses Students Researchers Office/Govt. The problem lies with existing materials that are not in digital form. Meanwhile there are technological problems to be resolved at the reader’s end. Even some producers of Hollywood films are storing all their new films in digital form so that they can be easily manipulated and transmitted electronically. thereby greatly increasing access to networks. The popularity of the Internet has spurred the computer industry to produce computers that will allow easy access to the Internet (including e-mail) in addition to word processing. Meanwhile. Students. Such computer systems could cost half of what they cost in the mid-1990s. This may increase motivation for self-directed learning. however. But libraries are converting their holdings into machine readable form (Path 1 2). their uses and their users (customers) are listed in Figure 21. There will. technologies may well change and become cheaper.6. be the problem of teachers having to learn about the potential of the new technologies made possible through telecommunications and having easy access to such technologies. college and university. They have to be converted through scanning (Box 6) that can be a manual process and hence slow and expensive. The screen is slower to read (20 30% slower). there are technical problems with the digital representation of materials available in the library.

Systems will be intelligent using techniques of AI (Artificial Intelligence). as for example in the simulation of conditions of environment or a model. the reader should remember that IT is notorious for its overoptimistic predictions especially the macro predictions of ‘megatrends’. ‘future shocks’. a country may restrict information as was the case in 1995 when the German government prohibited CompuServe from transmitting information with pornographic content. The linking may well be done by cable or telephone carriers and paid for by liberal depreciation allowances. what if the message were to cross national borders like from the liberal Netherlands to the conservative Britain? Should governments intervene by legislation or subsidize technology that gives parents and managers better control of ‘content’ being transmitted? These are questions that will eventually have to be addressed if telecommunications is to be free and unfettered. One attempt at this was the Telecommunications Act of 1996 in the US which is being contested in court. not just within an organization and corporation. other tax breaks and payment by advertisers.4 million downloads of material that had sexual content including pictures of humans having sex with animals. is the receiver responsible? Also. For the exchange of information across national borders we need the telecommunications infrastructure in other countries. This type of question raises many questions like: How does one balance openness with good taste? If pornography is allowed. Such issues of freedom of speech and expression across networks (national and global) are important and remain controversial as are the issues of security and privacy that remain unresolved. Even if the Act is considered legal. can we also allow say a manual on suicide? Who is responsible for slander and libel being transmitted on the network? Who is responsible for the content of what is transmitted? If a bookshop is not responsible for the content of all the books sold and the telephone carrier not responsible for the content of the telephone message. the question arises: can the US enforce its laws over the international Internet? Predictions often go wrong! In evaluating the predictions made above. Some countries will try to mix economic and information openness with authoritarian policies of restriction and control. libel or slander) is transmitted? Or is the sender responsible? Or. Systems will be more integrated. A taste of such problems arose in 1994 at the Carnegie-Mellon University where there were over 6. In the home. there will be use of animation. user friendly and multimedia. images and words. a geographic region and maybe even in an entire society. The mapping of the media to the end-user and its use will be determined by the end-user. CompuServe was then (December 1995) unable to control access to selected countries and so had to pull-off some 200 Internet sites from its approximately two million users. The superhighway will not raise many technological problems but social. ethical. but also make outsourcing of computing services more economically feasible. This should drop the cost of communications and increase not only corporate communications. Sometimes. digital. We will see computer technology and telecommunications merging information into a digital stream of sound.What lies ahead? vendor. Future systems will be more OLRT (OnLine-Real-Time) and interactive. The corporate network manager can assemble a network of bulk rates and offer internal services at a lower rate than the public provider. Notice that it is an infrastructure that will bring the links to the nodes rather than the nodes having to go looking for the link. In addition. moral and ethical ones. but also within a city. but in the long run free flowing information nurtures democracy as in Taiwan and Chile. This made the users in the US very irate for they felt that their freedom of expression was being controlled by another country. But there seems to be a distinct trends that future services will be more robust. Perhaps the most exciting and controversial development in the future of telecommunications and networking is the information superhighway which in the US is designed to connect every home. office and school with high-speed data and multimedia links. there will be two-way video terminals with multimedia capabilities and interactive services including movies and videoon-demand. a ‘leisure society’ and an ‘information 261 . but within the crosssubsidies allowed by governments. Such policies may enjoy short term success. is the carrier of transmission responsible for what (pornography. ‘third waves’. This can be very useful in both industrial and educational institutions. the user or the entertainment industry. the industries involved (especially the computer and entertainment industries) and the national government that have to provide the telecommunications infrastructure.

one should forgive those who are sceptical of the latest in IT predictions: the coming of the information highway. hardware ‘crashes’. ATM (Automated Teller Machines) and the use of credit cards. software glitches. In the US. colour TV took 20 years and cable television took 39 years before they became part of our daily lives. attending professional conferences and of evaluating IT as it may impact the organization.Telecommunications and networks society’. computer theft and the invasion of privacy? Will there be consumer acceptance and a willingness to pay for all the new additional services offered? The proponents of the information highway argue that all problems and obstacles will be overcome in good time. systems incompatabilities. at General Motors. They remind us that most revolutionary innovations took years before they 262 gained consumer acceptance. As examples. Why have our predictions for IT been so wrong. there were predictions that in the US there would be 250 000 robots or more by 1990 when a survey shows that in 1990 there were only 37 000 robots. only 11% of machine tools in the metalworking industry were NC (Numerically Controlled) and only 53% of the factories surveyed did not even have one automated machine. the average work-week was above 40 hours a week (including travel time) and the average worker in the US (and in many other countries) appeared to be working longer (and harder) than ever. Thus in manufacturing. 7). the 32 hour work week in the US resulting from computerization and automation by 1985 and a retirement age of 68 with more time for leisure predicted in 1967 is far from being a reality. paper consumption rose 320% over the last three decades. millions of dollars were lost in premature automation and robotization of the production line. We do not have the ‘electronic cottages’. the cost of failed software in the early 1990s is conservatively estimated at $900 million per year (Forester. 1991) estimated that 95% of the information in business enterprises is still in paper form and that only 1% of all information in the world is stored on computers. they recall that the radio took 11 years. and the extensive misuse and unauthorized access of electronic databases resulting in computer theft and the invasion of privacy. customer acceptance of small marginal changes in known products rather than large changes in new products. FMS (Flexible Manufacturing Systems) and MAP (Manufacturing Automation Protocol) have been slow to catch on and CIM (Computer Integrated Manufacturing) is still only a dream. One business strategy for avoiding or at least minimizing the danger of unexpected impact is to have an IT watcher. Predictions of a ‘cashless’ society and a ‘paperless’ office have been Utopian rather than realistic. Is this more hype or is it the coming of a multimedia revolution and the integration of computers and communications? Will the information highway offer interactive programming of over 500 channels and facilitate home-shopping? Who will take the risks involved and pay for all the telecommunications infrastructure necessary? Who will assure the consumer that there will be no open season for viruses. All State Insurance had the cost of its information system increase from $8 to $100 million and the time of completion extend from 1987 to 1993. the large number of successful ‘hackers’. In the computer applications area we also have been overoptimistic in our predictions. They argue that the real question is not whether we will have an information highway but when will we have it. In the 1990s. For example. The Bank of America abandoned one computer system after an investment of 5 years and $60 million. In the UK. Given that many IT predictions have gone wrong. Computers have permeated social and business life but have not transformed it as greatly as predicted. and are there any guidelines to help us prevent such errors in prediction? One problem is that we have often relied heavily on estimates made by the vested interests of vendors and inventors and on simple straight-line trend extrapolation and not paid adequate attention to human resistance. A study Business Week (June 3. In the US. In doing so. This may be a full-time person or someone assigned the duties of reading the literature. or even the ‘automated factories’ and ‘electronic offices’ that many had predicted for the 1990s. the ‘cashless society’. there has been a rise in the use of paper. Also. and the long time that it takes for IT to diffuse among customers and society. the technology watcher must . Despite the increase in the use of EFT (Electronic Fund Transfer). Another problem is that we do not always realistically estimate the complexity of information systems. It is also difficult to predict the unintended and unplanned consequences of computing such as the high level computer piracy. Some predictions are too long range to be evaluated today but the dates of some predictions have already passed without the predictions being achieved. 1992: p.

Another view of the evolution is to look at it as another stage in computing (Figure 21. . We can expect more innovations in technologies and their application in collaborative computing and plug-and-play computing. This evolution is shown in Figure 21. making related information all around the world available at the click of a mouse. Whether the message be in French or Arabic. fostering the illusion that all of the Net’s computers have been stitched together into one’ through hyperlinks. and. plug-and-play. management and end-users) consumer and organizational acceptance of innovation. assess (with the help of other analysts. This is shown in Figure 21. to personal computing in the 1980s.8). But there is a bigger problem. anticipate problems of implementation and the availability of adequate infrastructure. In either the technology view or the computing evolution view.7 Evolution of computing 263 .7. . where we are no longer concerned about compatibility of hardware and operating software or even programming languages. whether it be in text or pictures. give innovation time to diffuse and get consumer acceptance. We started with centralized computing in the 1960s. went to shared and distributed computing in the 1970s. Only by evaluating innovation in terms of both its technological potential and its human implications can an organization exploit IT and eliminate or at least reduce the consequences of incorrect predictions. and we had 5 GCS Non-Von Neumann Von Neumann Computing technology Electronic equipment M Ne ultim tw e or dia kin g Abacus ing ss ce g ro in e p ss dg roce le y ow el p log Kn rall hno Pa tec Stored AI N Program LA Program control Mechanization controlled of arithmetic operations Calculator 18th century 20th century Time Figure 21.8 with the corresponding hardware in use at the time. identify opportunities of potentially successful applications in the organization. Businesses around the world are waiting to use the ‘universal portal’ and use the Internet platform if only it were secure for monetary transactions of international commerce. as long as it is enveloped in a simple language like HTML it will be recognized by all computer systems around the world as long as they are connected to the Internet. managing Summary and conclusions Networking in telecommunications can be seen as a recent stage in the evolution of computer technology. that of financing it. we have not reached the end. The Internet in recent years has been called the ‘throbbing new center of the computing universe . evaluate the sources of information.What lies ahead? distinguish between technological forecasts and market trends. finally. cooperative processing and network computing in the 1990s.

machine translation. a freewheeling on-line public discussion place. The use of computers in many offices and homes around the world may never be the same. libraries. What is currently in use cautiously or experimentally may soon become necessary and a way of life. software for program navigation. the Internet may lose some of its character of being an electronic democracy. as well as efficient and friendly end-user computing. high resolution flat screens. joint ventures. which includes technologies (like digital telecommunications). And doing much of this in cyberspace. playground and a social club all at once. and telephone (which is two-way but low capacity) subscribers. acquisitions.8 Evolution in computing it and controlling it once the demand increases and the financial support from America dries up. intelligent peripherals and secure user identification. content programming (including education and movies). sound. This deployment of multimedia technology will be on a platform of interactive processing. continuous speech recognition. and even travelling to far-off 264 places by virtual reality. the future may well see our shopping. this will include the common use of hand-held computers connected by international networks to anyone around the world. In terms of applications. of a workplace. Access of information will be to businesses.Telecommunications and networks Intranet. very large client base such as television (including cable in the US. educational and training programs. use of natural language for computer interfaces. Such conglomerates are designed to bring within one firm’s jurisdiction the necessary informational infrastructure. individuals. capacity (including satellite and fibre optics transmission) and switching abilities to deliver full information services to all customers. or by hostile takeovers. speech synthesis. From a technological point of view. using EFT (electronic fund transfer) instead of cash. graphics and animation. using e-mail instead of visiting friends. Furthermore. replacing the library visits by using an interactive videotext. substituting distance learning and digital libraries for education and training at fixed sites. Access to the convergence of computer. Java Internet 1990s Plug-and-play Collaborative computing Network computing Client−Server PCs and workstations 1980s Personal computing 1970s Minis Shared computing Distributed computing 1960s Mainframe Centralized Figure 21. video. This may be the time when we say good-byte to the freedom of the flat fee rate and the downloading of free software and unrestricted and fast e-mail. The future of IT will see much integration. which is one-way and high capacity). Communications will take place in cyberspace on an electronic highway initially with an NII (National Information Infrastructure) and later on an III (International Information Infrastructure). banking and reading the newspaper from our computer at home. The other dimension of integration is largely organizational and may come from strategic partnerships. this access may soon become global. communication structure and . The integration will be partly technological: the integration of text. In the resulting reorganization. and databases.

Some of the goals of the fifth and sixth generation computers may prove to be technologically unfeasible. which up until then had been buried in the cabinets. as well as enhance the transborder flow of data and information. we may well edge towards a cashless and paperless society. then there will most likely be regulation by governments. they will have a prominent role in shaping our telematic society. lack of markets.What lies ahead? programming media will no longer depend on location but can take place anywhere. legal restrictions and user resistance are factors that may affect the speed and direction of change. pp. (Montague and Snyder. 1995. Limited resources. Case 21. ‘the digital city has become a model virtual community for the digitally and politically active. high risk of capital investments. The speed of technological and organizational changes will require adjustment of consumer behaviour and still possibly be pro-competitive. but rather that we must understand the true nature of this latest of man’s inventions and learn how its powers can be combined with our own abilities to be used to the best advantage for humanity. Founded in 1993. In any event. universal service to consumers. Computing will not only be democratized but it will offer the consumer more choices and will be end-user friendly. unfavourable tax policies).1: Amsterdam’s digital city The digital city of Amsterdam is a city within a city replete with caf´s. and may be even in a visual environment and in virtual reality. provide open access to corporate players. kiosks. Planning must begin now for the computerized society to come. from one of many devices. government regulatory schemes or interference (for example. leisure. The idea was to shed some light on the minutes of city council meetings and official policy papers. The author hopes that this book will help you to prepare for the challenges and opportunities ahead. product development priorities in other fields. and promotion of cultural and linguistic diversity in traffic. Along the way.’ Source: International Herald Tribune. Also desirable would be that national governments provide ramps to the information superhighway and enforce international standards for the protection of intellectual property and data. 1 2). Case 21. Early spamming was when advertisers sent thousands of advertisements to addresses that did not want those messages.’ Since business and IT managers often serve in a leadership role in their communities as well as at work. regional (like the governmental agencies technopolis strategy) and private networks with governments agreeing to the interoperability of national networks. Not all the predictions made in this chapter will come to pass. home and community. 9. e billboards. we must be aware of the caution by the Librarian Emeritus of the US Congress Library. political party platforms.2: Spamming on the Internet Spamming is the flooding e-mail on the Internet with undesired and unauthorized information. our t´l´matique ee future will require a redefinition of work. As stated by Montague and Snyder: Many deplore the computer and some even fear it as more monster than machine. If there are important restraints (actual or perceived) to competition. town-squares. 1972. However. be it about local or national elections. inspiring analogous projects across the Netherlands and Europe’ and has a ‘population’ of 30 000. This does not imply resignation. Daniel Boorstin: ‘We have created and mastered machines before realizing how they may master us. 15. Oct. This type of spamming was controlled by the informal ‘netiquette’ on the Internet and was policed by members who spammed the advertiser to the point that the 265 . an information centre and offices. we must adjust to it. p. house rentals. A high rate of changes in our information age and their many reasons and dismaying consequences confuse our ability to predict the future. The city’s manifesto includes the simple philosophy: ‘Every plugged-in Amsterdammer should be able to click into the political domain to receive and exchange information on the latest governmental developments. however. We may well see a blend of national. or interestgroup agendas. Whatever we think of it. or too expensive to implement.

Surely there will be a solution to such spamming problems but just as surely there may well be other twists to spamming the interruption of a valuable e-mail service rendered on the Internet. Mr ElmerDewitt painstakingly unsubscribed himself from 106 of these mailing lists only to find that the next day he was on 1700 more lists. p. But meanwhile.’ It is expected that France Telecom will be able to meet the objective by routing Internet traffic along the same Tanspac network developed for the Minitel. Digital Corporation’s Internet Technology Manager reminds us: ‘The Minitel is both a brake and a boon. Every innovative application of telecommunication seems to pose a problems that calls for an innovative response. the princ cipal objective is ‘to make access to the Internet possible for all French citizens. users were able to access stockmarket prices and perform banking transactions both nationally and internationally. Henri Gourald. Source: Time. 56. It was unable to sell its system to other countries. France seemed poised to take the lead in the race of developing the Information Society but never capitalized on its technology. Besides teleshopping. the Internet grew 50 000 times faster in the US than in France. pp. Between 1992 and 1995. Analysts in Paris calculated that it would be cheaper to give every household a free computer terminal with real-time access to the telephone directory than to annually update the printed telephone directory. with the political will to introduce a conceptually advanced technology. The White House called in the Secret Service that managed to reduce the flow to 1200 messages a day. however. as well as a monopolistic PT&T for all telecommunications. 77. Source: International Herald Tribune.5 million homes with Minitel terminals to be connected to the nationwide teletext system. ‘We tried’. Minister of Technology. Jan.Telecommunications and networks Internet provider for the advertiser would cut the advertiser off the Internet. Minitel’s hardware is slow and monochrome text-based with rudimentary graphics as compared to the faster coloured multimedia ability to click around the world afforded by the Internet. Senior Editor of Time magazine.’ Today. 1. France has concentrated on local services which included a credit-card reader that makes it easy and safe for teleshopping because the verification system is at the terminal and does not have to be sent down a telecommunications line. This crowds legitimate e-mail messages and is a great inconvenience.’ on the Internet. One such targeted person was Elmer-Dewitt. France introduced Minitel terminals in Paris and other selected regions. According to Fran¸ois Fillon. A new twist to spamming is where people are targeted largely for political purposes by being put on mailing lists which then spam the targeted person with unexpected and unwanted messages. p. October 2. Case 21. 1996. 1996. He found himself being put on 106 mailing lists that generated about 50 messages a day each.3: Minitel its future its past and Around 1980. at a price which is attractive and the same anywhere in France.’ These ingredients were a centralized government. The system also has the ‘Minitel Rose’ which is the equivalent of ‘alt. The Time editor is not the only one spammed. 8. the price of the terminal went up and other related investments came to over 56 billion francs. spokesman. France has now recognized the potential of the Internet and is planning to catch up. The Minitel was connected to message-switching network called the Transpac which was integrated to the telephone system. said the France Telecom 266 . Terminals will still come free or at a nominal rent and access to databases like a newspaper database will be charged at around 9 francs a minute and the bill simply added to the telephone account retaining the simplicity and reliability of the Minitel system. ‘But we could never find the same ingredients in other countries that we had here. The President of the US was spammed at his White-House e-mail address. Don’t forget that it has installed an electronic commerce ethos in 25% of French households. 6. There were 25 000 providers of on-line services with 2 billion calls logging over 110 million hours of connect-time. Over the years. and Information Week. Minitel’s technology design was rooted in the 1980s though Minitel’s concept of letting the program and data reside in cyberspace and offering end-users a cheap terminal interface was conceptually advanced compared to other approaches at the time. the Internet was growing beyond anyone’s expectations and France was being left behind. March 18. This allowed over 6.

3 billion 150 million 278 million 132 million 230 million 18 million 47 million Source: Business Week. 1994. Y. Brody. The Computerization of Society. (1987). Levy. Heilmeier. H. Time. N. PC World. 45 51. 18(11). and Minc. Coleman. Keyes. Supplement on ‘The internetwork decade. Hansson. R. 98(3). 13(9).) (1991).D. 142 154. The AI market in the year 2000. M. Nash. 34 44. Wired. (1996). Forester. Knorr.H. Huws. AI Expert. E-money: that’s what I want. 75 78. Carlyle. (1991). Evolution of intelligent network concepts. 2 11. (eds. S2 S29. AI Expert. 213 215. Artificial intelligence in management: future challenges. Datamation. 793 809. Computer Communications. May 1995. Where the hackers meet the rockers. Galliers. 10 13. J. 25 31. Man and the Computer. S. special Issue on 21st Century Capitalism.J. Evans. (1993). Megatrends or megamistakes? Whatever happened to the information society? Computers and Society. The next insanely great thing. AI on a chip. (1994). 6(4). Futures.What lies ahead? Supplement 21. 475 502. Brancheau. (1993). J (1991). 16(1). wired world. 135 157. S. 15(4). Lowry. 4(2). Economist. Mujjender. Key information systems management issues for the 1990s. 30 34. 178 180.4 billion 34 million 1. MIS Quarterly. 45(7). 1(4). 194.’ Data Communications. E. AI Magazine. (1992). Spring. A and Snyder. J. 8(1). R. 134 143. D. Jobs. (1972). 328(7828). (1993). IEEE Communications Magazine. 80ff.2: World-wide predications for 2010 compared to 1994 1994 Wired telephone lines: Wireless telephone lines: Number of PCs: Number of desktop computers: Number of mobile computers: 2010 607 million 1. union and technology. Towards 2017. pp. S. (1993). The Journal of Information Systems. Niederman. 2(12). J. 22(1 4). U. 33(18). Fifth generation computer systems. (1995). MIT Press. and Herman. T. M.1: Percentage growth of phone lines connected to digital exchanges in the 1990s in Europe Country France Germany Italy Netherlands Spain Sweden UK Rest of Europe Western Europe End of 1994 88 40 64 62 50 63 81 53 62 End of 1999 (predicated) 100 73 93 84 74 96 88 74 86 Source: Dataquest.O. G. 33 38. (1994). Transactions on Knowledge and Data Engineering. Information systems management issues for the 1990s. Telematics India.C. (1991). Wired. November. 24 31. (1980). 174 177. Supplement 21. p. Montague. The death of telephony. art. 3(2). State of the market. (1994). 128(7). Journal of Systems Management. J. Technology Review. 1 162. Special issue on ‘1994 Information Technology Guide’. Malhotra. (1995). (1991). Software engineering in the twenty-first century.R. Telework: projections. Strategic technology for the next ten years and beyond. 125 159. (1992). A. Software’s nest ware: putting the user first. Motiwala.. 12(3). pp. S. Internet@Crossroads. G. 8(1). 12 17. A. 8(1). 31(3). AI Expert. F. sJ. Controlling copyright infringements of intellectual property: Part 2. (1992). (1989). and Wetherbe. Lillis. Bibliography Brusseis. It’s a wired. Gilder. K. 23(1). Computing Now!. (1995). 102 107. (1990). Auerbach. Nora. 158 163. 267 . J. 71 87. Fortune.$$$. J.

128(7). Technologies in the 1990s. Customers will be the innovators.D. A. J. S. 91 99. Y. A look into the 21st century: people. Special Issue on ‘Technology. 14(1). 30(2). Toffler. Telecommunications. (1989). J. the Control of the Twenty-First Century. 105 107. P. Knopf. 12(2). 4(1).J. L. Seitz. 1995’. E. pp. C. Expert Systems with Applications. Rockhart. Forecasting and the Myth of Rapid Change. Tatsuno. Artificial neural networks: an emerging new technique. Creating a winning culture. 26 51. (1993). Spectrum. Technopology: Technology and the Surrender of Culture. (1989). H. Postman. (1991).R. Computers then and now Part 2. (1986). The Third Wave. judgement.L. Fortune. Information Age. 66 90. Personal Computing. Summit Books. and Short. K. 58(2). Prentice-Hall. Megamistakes. (1995).E. N. (1992). Collier Macmillan. (1995). and Watkins. 37 39. (1980). Schroth. Schnaars. and wisdom. and Peterson. M. A. R. Proceedings of the 1996 Computer Conference. Ten major trends in strategic networking. Spence. Database. (1996). Siemens Review. (1993). Rheingold. 7 12. 32(1). (1992). 1 10. High Technology.Telecommunications and networks Reed. and Mui.F. S.R. Wilkes. Beyond expert systems reasoning. Yoon. 115 119.V. 33 42. M. William Murrow. 55 58. 23(1). business and computer. (1990). S. von Simpson. (1991). Rowe. IT in the 1990s: managing organizational independence. (1990). Virtual Reality: The Revolution of Computer Generated Artificial World and How it Promises and Threatens to Transform Business and Society. Technopolis Strategy: Japan. 268 . Management Review. 29(10).

Sheppard. token. Some terms are duplicates even in telecommunications like MHS for Message Handling Service and for a product by Novell. To avoid confusion. There are also books on acronyms alone. flag. There are a number of books and dictionaries on computing which include telecommunications and networking terms. bonding. open. and ATM does not mean Automatic Teller Machine but Asynchronous Transfer Mode. There are also terms that are downright confusing because they have already been used in the computing vocabulary. master slave. Some of these are innocent looking words used in daily life but mean something different in networking like backbone. bus. 269 . platform. a telecommunications product vendor. All are defined in this glossary. Similar words that are confusing are T-1 and T1. peer. most of these terms are defined in the text. Towell and Helen E. bridge. like the 1986 oversized book Computer & Telecommunication Acronyms by Julie E. Java layer. transparent and virtual. like encapsulation in OO (objectoriented) methodology and in networking. Examples are SDLC which does not mean Systems Development Life Cycle but Synchronous Data Link Control. Some words have different meanings in computer science.GLOSSARY OF ACRONYMS AND TERMS IN TELECOMMUNICATIONS AND NETWORKING Computing is notorious for its jargon and acronyms. Telecommunications and networking should take more than their share of the blame. host. They have introduced many new terms and acronyms to the computing vocabulary. robust. packet. cloud.

IP Internet Protocol. FLOPS Floating-Point Operations per Second. IXC Inter eXchange Carriers. FAQ Frequently Asked Questions. IPX Protocol for transmitting and moving information over a network. ISDN Integrated Services Digital Network. ee CIX Commercial Internet Exchange. BISDN Broad-based ISDN. ECMA European Computer Manufacturers Association. ITU International Telecommunications Union. CSMA Carrier-Sense Multiple Access. BBS Bulletin Board System. 270 EDH Electronic Data Handling. HTML HyperText Markup Language. Hz Hertz. CNMA Communications Network Manufacturing Association. CAT Computerized Axial Tomography. ISP Internet service Provider. API Applications Programming Interface. now ITU. ICMP Internet Control Message Protocol. BSI British Standards Institute. 2nd generation. BRI Basic Rate Interface. FDDI Fibre Distributed Data Interface. Commerce and Transportation. DIN Name of the National Standards Organization in Germany. ETSI European Telecommunications Standards Institute. CEPT Conference European des Administrations des Postes et des T´l´communications. E-mail electronic mail. cycles per second. IEEE Institute of Electrical and Electronics Engineers. ATM Asynchronous Transfer Mode. ITU-T ITU-Telecommunications (standards). APPN Advanced Peer-to-Peer Networking. HTTP HyperText Transfer Protocol. EUF European ISDN Users Forum. IEC International Electrical Technical Committee. . CD-ROM Compact disk for read-only-memory. FTP File Transfer Protocol. CCIR International Radio Consultative Committee. EIUF European Computer Manufacturers Association.ACRONYMS IN TELEPROCESSING AND NETWORKING ANSI American National Standards Institute. CDDT Copper Distributed Data Exchange. CEN European Committee on Standards. CBDS Connectionless Broadband Data Service (SDMS in the US). EDIFACT EDI for Administration. ASCII American Standard Code for Information Exchange. CCITT International Telegraph and Telephone Consultative Committee. ISO International Organization for Standardization (in Switzerland). B channel Barrier Channel. EDI Electronic Data Interchange. ESPIRIT European Strategic Programme for Research and Development in Information Technology. CDPD Cellular Digital Packet Data. CENELEC European Committee for Electromechanical Standardization. B-ISDN see BISDN. CT-2 Cordless Telephone. GUI Graphical User Interface. EuroCAIRN European Cooperation for Academic and Industrial Research Networking. CDMA Code Division Multiple Access.

TTC Telecommunications Technology Committee (in Japan). V. X. PCS Personal Communications Service. SLIP Serial Line Internet Protocol. NOS Network Operating system. LAN Local Area Network. MRI Magnetic Resonance Imaging. SNA Systems Network Architecture. also referred to as a ‘home computer’. PTT Poste de T´l´phony (and) T´l´graph. SMDS Switched Multimegabit Data Service (same as CBDS). NII National Information Infrastructure (in US). SMR Specialized Mobile Radio. PAL Phase Alternating Line. URL Uniform Resource Locator. 271 . X. Each for a different standard. TCP/IP Transmission Control Protocol/Internet Protocol. LEC Local Exchange Carriers. Web short for WWW. SNMP Simple Network Management Protocol. The xx are numerals. TERENA Trans-European Research and Education Networking Association. MAN Metropolitan Area Network.Acronyms in teleprocessing and networking JTC1 Joint Technical Committee 1. PRI Prime Rate Interface.11 The dominant windowing system on Internet. MPEG Motion Picture Experts Group. PRI Primary Rate Interface. Mbps Megabits per second.7366. LEO Low Earth Orbit satellite. MIPS Millions of instructions per second. SMTP Simple Mail Transfer Protocol. SONET Synchronous Optical Network.25 A packet switching protocol defined by CCITT. T1 Transmission link for distances up to 50 miles at 9. WAIS Wide Area Information Service. WWW World Wide Web. PBX Private Branch Exchange. VERONICA Very Easy Rodent-Oriented Net-wide Index to Computer Archives. MHS Mail Handling Systems. Kbps Kilobits per second. X. RPOA Regional Private Operating Agency. WAN Wide Area Network. OSI model Open Systems Interconnection model developed by the international organization for standards (ISO). PC Personal Computer. PoP Point of Presence. RACE R&D in Advanced Computer Technology (in Europe). SDLC Synchronous Data Link Control. ee ee PVC Permanent Virtual Connection.xx Designation for standards in the field of integrated circuit equipment. SGML Standard Generalized Markup Language.544 mbps T1 Committee on Standards in the US. LINX London INternet eXchange.400 A common protocol for standard mail messaging. PCN Personal Communication Network. SVC Switched Virtual Connection.6 to 1. PTM Packet Transfer Mode (by IBM). PnP Plug-and-Play. RBOC Regional Bell Operating Companies. It is also the European name for SONET in America. SIO Scientific and Industrial Organizations. SDH Synchronous Digital Hierarchy. PPP Point-to-Point protocol. PIPEX Public IP Exchange. MIME Multipurpose Internet Mail Extensions. STM Synchronous Transfer Mode. SDH A framing format chosen by B-ISDN. NNTP Network News Transfer Protocol. T3 Transmission link up to 500 miles at speeds up to 44.

Bitnet is a subset of the Internet. the bandwidth would be 4000 Hz for a range of 400 to 4400 Hz). bus configuration in which all nodes are connected to one main connection line. backbone network a central network to which other networks connect. and named after Emile Baud.g.GLOSSARY access provider a company that sells access to the Internet. broadband a method of transmitting large amounts of data. cluster addresses represents a domain as a single address rather than a set of individual addresses. browser a program that allows you to download and display documents from the World Wide Web. byte 8 bits. . cloud boundary of a packet switching service. client it is a user system accessing services as in a ‘client server-system’. carrier a public transmission system in the US and Canada corresponding to the PT&T in Europe. Archie A system on the Internet that allows searching of files on public servers by autonomous FTP. agent programming code designed to handle background tasks and perform actions when a specific event occurs. analogue a transfer method that uses continuously variable physical quantities for transmitting data and voice signals over conventional lines. channel a path between sender and receiver that carries a stream of data. cable TV a broadcasting system that uses giant antennae. codec short of coder/decoder in multimedia and is the equivalent of a modem for non-multimedia transmission. bandwidth the information carrying capacity of a telecommunications media. A circuit switched digital channel that sends and receives voice and data signals at speeds of 64 kbps. also a pathway between two computers or between computer and control unit or devices. therefore not having a fixed timing relationship. bulletin board a system that allows users to post messages and receive replies electronically like on the Internet of an information service provider. access time interval of time between the instant at which a call for data is initiated and at which it is delivered. asynchronous not derived from the same clock. baud rate the speed rate of a data channel expressed at bits/second. Often used as a symbol for a network. brouter a bridge and a router. cable a flexible metal of glass wire or group of wires. bandwidth-on-demand contract for bandwith as it is needed not a contract for fixed capacity (see SMDS in US and CBDS in Europe). cell term used in switches D 8 octets. asynchronous transfer mode a packet oriented transfer protocol that is asynchronous. see daemon. backbone the top level in a hierarchical network. a telecommunications pioneer. B-channel Bearer channel. Stub and transit networks which connect to the same backbone are guaranteed to be interconnected. 272 baud a measure the speed with which a moderm transmits data. backplane a pathway in which electrical signals travel between devices conceptually similar to a bus. also the upper and lower limits of a frequency range available for transmission (e. beta as in a beta test is the preliminary testing stage.

also called a processor. hypermedia video and text files transmitted by way of the Web. connection-less service a type of service in which no particular path is established for the transfer of information. and languages of each other. cybersex. finger a service that provides data about users logged on the local system or on a remote system and used for security purposes. dedicated line same as leased line. daemon a program that runs in the background on a Unix workstation waiting to handle requests. a bounded short-term relationship (switched connection). Gopher a tool on the Internet that searches fields of a hierarchical nature from a menu. dark filtering part of bandwidth held in reserve. greater speed and more flexibility in transmission than analog. hub a central switching device for communication in a store-and-forward mechanism. end-to-end delay the time span between the generation a data unit at its origin and its presentation a the destination. hacker describes a skilled programmer who has a mischievous bent and likely to intrude into computer files without authority to do so. configuration shape. a large computer that serves other computers or peripherals. D channel a digital channel that carries control signals and customer data in a packet switched mode. connection-oriented service a type of service in which. for any given call or session. a science eviction writer to represent the counterculture outlaws who survived on the edge of the information highway. etc. cyber see cyberspace and cyberpunk and then deduce the meaning of cyberculture. Dante is an organization working towards a seamless integration of all Europe’s networks. home-page an explanation of the database on the Internet and may include a description of the content and an explanation of how to access and use the database on the server.Glossary compatibility the degree to which there is an understanding of the same commands. e-mail electronically transmitted messages. Demon a UK-based Internet access provider. It is usually an unattended process initiated at startup. See agent. encapsulation inserting a frame header and data from a higher level protocol into a data frame of a lower level protocol. an organizational area. Ethernet a network cabling and signalling scheme using a bus architecture. icons graphical symbols that represents a function or a subroutine. frame relay a networking technology that exploits the high quality fibre optics to deliver data up to 10 times faster than today’s packet switching. dot looks like a period but is used in a telecommunications address as a separator of the address subfields. related documents elsewhere in a collection. contention arises when two or more devices attempt to use a single resource at any one time. offer from larger to smaller computers. gigabyte a billion bytes. arrangement or parts. Dante’s services complement those provided by the national research networks in Europe. cyberspace a term coined by William Gibson. 273 . Digital transmission lines offer faster speeds. information traverses only one path from sender to receiver. gateway a communication program (or device) which passes data between networks having similar functions but dissimilar implementations. cyberpunk coined by Gardnier Dozois to evoke the combination of anarchy and high tech. cyberworld. It can regenerate signals as well as monitor signals. filtering elimination of unwanted network traffic. hypertext a link between one document and another. digital the representation of data in on/off signals of 0’s and 1’s. engine the portion of a program that determines how the program manages and manipulates data. or a boundary-less connection (connection-less). connection type specifies whether an application has a long-term relationship (permanent connection). domain a part of naming hierarchy on the Internet. format. download transfer of data. firewall software that controls network traffic through a node. host a system that has at least one Internet address associated with it.

in telecommunications it is a unit of data transmitted. looping a condition in packet switching networks when packets are travelling around in a circle. latency the time interval between the instant at which an instruction control unit initiates a call for data and the instant at which the actual transfer of the data starts. audio. Any program written in Java can be executed in any computer or digital device ranging from a machine tool to a computer. leased line a private line for dedicated access to a network. Mosaic Hypermedia browser on the Internet that allows searching using hypertext and a GUI. information highway a term used in the US for communications across telecommunications networks to transfer information.g. isochronous time dependent. message in information theory it is the ordered series of characters intended to convey information. usually accidentally. LINX is an organization set up to provide interconnectivity for UK Internet service 274 providers and also to further the cause of the UK within Europe. master primary and controlling unit. multilink a software technique of adding channels to networks. layer a group of services that is complete from a conceptual point of view. text. animated graphics or full motion video. megabits just over 1 million bits of binary digits (0 or 1). network An arrangement of nodes and connecting branches. and extends across all systems that conform to the architecture. newsgroup these are bulletin boards on the Internet. e. Java is a programming language designed for network computing. real-time video or telemetry data. also referred to as private line or dedicated line. Graphical User Interface. modem a device that allows a computer to transmit information over a telephone line. kluge a clumsy but serviceable solution. The slave responds to the master’s commands. that is one of a set of hierarchically agreed arranged groups. mung to destroy data.Telecommunications and networks infobahn the European version of the ‘information highway’. packet switches small computers linked to form a network. packet a unit of information travelling as a whole from one device to another on a network. link a line channel or circuit over which data is transmitted. peer a functional unit that is on the same protocol layer as another. mail gateway a machine that connects two or more electronic mail systems and transfers messages between them. See access time. A modem performs the function of a modulator and demodulator. open systems ability to connect any two systems that conform to a reference model and its associated standards. multimedia information that may be in one or more forms including data. Local Area Network (LAN) a network offering connection services within a very limited area typically the same building or campus. octet 8 bits. open protocols protocols that do not purposefully favour any single manufacturer. information infrastructure a publicly owned facility like the roads or electricity utilities to be shared and used by others infrastructure. JPEG a way of compressing still images and video which is widely used on the Internet. software and user. peer-to-peer communication communication in which both sides have equal responsibility . also called an information. master slave a communication in which one (the master) initiates and controls the session. interface connection and interaction between hardware. peer-to-peer a network architecture where a user’s PC doubles as a server rather than accessing centralized file or print servers. killer app short for killer application which is a very successful computer application of computers. intranet a private network using the protocols and infrastructure of the Internet. jitter the variation a packet may get through a service. graphics. Metropolitan Area Network (MAN) a network that connects computers within a metropolitan city area.

token ring a signalling device where a special message. synchronous derived from the same clock. pixel a smallest element on a video display screen. token a series of bits which when grabbed by a user allows the use by sending packets across the network. propagation delay a delay in the transmission of information from source to destination. virtual pertaining to a functional unit that appears to be real but whose functions are accomplished any other means. pix(picture) element. service provider an on-line service that lets users connect to the Internet and which in turn is a gateway to the Internet. surfing exploring the Internet. video-on-demand an interactive system that allows you to point your remote control at the screen and select a desired program. therefore having a fixed timing relationship. private line same as leased line. teleprocessing is a computer-supported technique for providing a number of remote users access to a computer system. Veronica an index to computerized archives. Pixel is short for picture element. platform the principles on which an operating system is based. seamless smooth without awkward transitions. gives a node permission to enter a message or frame into the ring. Telnet a tool for interaction communication with remote computers. standard describes how things should be. All real-time is on-line. robust refers to a solid program that works properly under all normal but not abnormal conditions. Of no significance. In OSI terminology. processor see engine. server a central computer which makes services and data available. but not all on-line is real-time resource allocation a mechanism to allocate resources according to a promised resource reservation. upload opposite of download. plug-and-play usually referred to for hardware that can be attached (plugged-in) almost anywhere and starts operating (playing) without the need of special interfaces. pel short for pixel. pipeline allows for simultaneous or parallel processing within a computer. switching means of relaying information from one path to another. terrestrial earthbound. sniffer synonymous with network analyser and used for diagnostics. 275 . seamlessly blending smoothly. sounding a hardware technique of adding channels to networks (same as bandwidth-on-demand). protocol sets of rules and agreements (on format and procedures). peer-to-peer networking is where files could be exchanged and terminal sessions established on a non-hierarchical basis. T-1 carrier this system uses a time-division multiplexing to carry voice channels. switching system consists of hardware and software. a router is a network layer intermediate system. passed from node to node. scalability capability of being changes in size and configuration. real-time processing with updated database. portal a meeting point between local and long-distance services. It could be a dot and is sometimes called a pel. photonic switch device that switches optically rather than convert signals to an electronic path as in conventional semiconductor technology.Glossary for initiating a session compared to a master-and-slave relationship. roamer a subscriber to a telecom in locations remote from the home-service area. transparent virtually ‘‘invisible’’. a switching system’s primary purpose is to form dynamic connections between channels. router a system responsible for making decisions about which of several paths traffic will follow. posting a message on the bulletin board. a system responsible for selecting the path for the flow of data from among many alternative paths. Universal Resource Locator a URL is the technical name of the World Wide Web address. transceiver a physical device that connects a host to a LAN. switched 56 digital service are 56 kbps provided by local telephone companies and long distance carriers.

Wide Area Network (WAN) a network that serves a geographic area larger than a city or metropolitan area.Telecommunications and networks voice line a communications link usually limited to transmitting data at the bandwidth of the human voice. 276 . Web short for World Wide Web. Web site a collection of files on the web built around a common subject or theme. voice mail a service that stores voice messages for users and enables them to retrieve and hear their messages in various ways.

40 1 Configuration management 143 Cooperative processing 194 5 Cordless 29 30 Cyberspace 232 Digital library 208 9 Digital money 179 80 Digital world 251 Distance learning 208 9 Distributed data processing 15 6 Distributed multimedia 199 200. 171 84 Global networks and businesses 179 80 Global outsourcing 175 6 Hardware for network manager 147 8. 172 Intelligent devices 43 5. 203 4 Client server paradigm 100 12 Client server system’s impact 106 8 CMIP 146 Comm software 156 Communications security 128 30 Compression 38. 56 Film-on-demand 206 Firewall 238 Frame relay 55 Frames 39 FTP 239 Future of telecommunications 21 3. 224. 251. 205. 151 FDDI 46. 255 6 ATM 62 7.INDEX Accounting management [of telecommunications and networks] 143 4 Addressing 41 2 Addresses on the internet 222 Agents 252 ANSI 117 Applets 252 Applications 6 10. 146. 196 7 Electronic money 179 80 Electronic publishing 213. 187 248 APPN 82 5 ARPANET 11. 251 277 . 133 4 B-ISDN 72 3. 120. 75 CCITT 41. 196. 210 11 E-mail 221 4 EDI 188 9. 239 EFT 190 4. 34 Global network in developing countries 174 5 Global network infrastructure 172 Global network’s impact 173 5. 259 Encryption 128 9 End-user friendly systems 251 Ethernet 55 ETSI 118 20 European Standards Organisation 115 20 Fault management 141 2. 241 68. 242. 225 Cellular phones 29 Cellular radio 29 CERN 255 Channels of transmission 17 9 Circuit switching 254 Client 102 3. 180 1 Global networks 21. 55. 252 Biometric systems 125 Bridge 38 9. 255 Gateway 38 Glass fibre 26. 121 Bandwidth management 61 2 Bandwidth-on-demand 62. 237 Information service providers 225 Information services on the internet 241 Infrastructure 161 6. 48. 153 4 Home banking 193 Home-page 234 HTML 234 HTTP 234 5 Hubs 45 6. 259 60 Authorization controls 126 8. 106. 117 18. 251 3. 251 Hypermedia 211 Hypertext 211. GAN 211.

251 Standards 112 22. 251 Levels of organization in network management 91 4 Logic bomb 128 MAN 59. 117 20 Standards organisations-American 111 Standards organisation European 118 20 Standards organisation Japaneese 120 Standardization 189 90. 69 77. 257 8 ISO 113 8. 257 Surfing on the internet 234 5 Switches 38 Switching 38 47. 230 48 Internet connections 232 4 Internet organization 240 1 Internet security 230 40 Intranet 11. 82 6. 78 80 SNMP 146 Software for network management 144. 203 Java 11. 251 Interfaces 19 20 Internet 218. 252 IP 45 ISDN 21 2. 151 Processing in client-server systems 105 6 Protection of intellectual property 178 9 Protocols 45. Message handling systems 165 95 Messaging 185 97 MHS 195 6 Microwave transmission 28 Middleware 157 8 Modem 42 3 Monitor on traffic 141 MPEG 41. 155 6 Open systems 80 7. 65 6. 116 7 Organization for networking 91 9 Organizational levels for network management 91 4 OSI model 80 7. overview of 15 24 NII in the US 163 6 NII issues 167 70 NIIs around the world 162 3 NOS 142. 252. 204 MPR 40 Multimedia 187 97 Multiple protocols 84 5 National infrastructure 161 70 Network administration 93 5 Network chanracteristics 50 Network management 91 5. 39. 252 Standards organizations 111 5. 67 Satellite 28 9 Security 123 39 Security management 142 3. 146 Telecommunications. 48 50. 151 Security on the interent 236 40 Servers 100 1. 153 8 Software piracy 178 9 SONET 56. 151 Personnel for network management 91 4. 54 Public networks 253 Regulations 252 Remote processing 16 Repeater 39 Resources required for teleprocessing 153 60 Roamer 29 Router 21. 148 51 Planning of telecomm. 53 6 Switching management 62 TCP 45 TCP/IP 45. 202 3 Smart cards 193 4 Smart devices 43 5 SMDS 252 SNA 11. 104 5. 140 52 Network security planning 132 3 Network systems architecture 78 88 Networks. 204.Index Interconnectivity 20 1. 173. and networks 95 8 Plug-and-play 45 Poller on traffic 141 Predictions go wrong 261 3 Private networks 252 Problem management 141 2. 251. 252 PDA 30. 252 Knowbots 252 LAN 48 57. 56 PAD 30 1 Parallel processing 153 5 Passwords 125 6 PCS 31 2. 251 Performance management 142. 41. 225 6 278 . 147 Software for telecomm. 116 7 Packet switching 254 Packet-radio 29 30 Packets 39. overview of 15 24 Telecommuters 216 21.

56. 52 5 Transborder flow 176 7 Transceiver 29 Transmission 26 37.Index Teleconferencing 188 Telematique society 251. 56 7 Wiring 26 8 Worm 130 WTO 260 X-25 54. 257 Telemedicine 205 7 Teleworkers 220 1 Telmatic society 251. 224 279 . 32 3 Transmission media 165 6 Trojan horse 128 Video-on-demand 206 Videoconferencing 204 6 Viruses 128 32 WAN 59 61 Wired society 258 Wireless 29 30. 257 Terminal use controls 124 5 Threats and counter threats to secure telecomm. 136 Token ring 55 Topologies of networks 16.